Terradev CLI Command Reference v5.6.4
Generated from terradev --help and subcommand --help output.
terradev
Usage: terradev [OPTIONS] COMMAND [ARGS]...
Terradev CLI - Cross-Cloud Compute Optimization Platform
Parallel provisioning and orchestration for cross-cloud cost optimization.
Save 30% on end-to-end compute provisioning costs with real-time cloud
arbitrage.
Options:
--version Show the version and exit.
-c, --config TEXT Configuration file path
-v, --verbose Verbose output
--skip-onboarding Skip first-time setup
--help Show this message and exit.
Commands:
agent Provision and manage heterogeneous agent fleets.
agentic-serving Agentic inference serving KV cache TTL, prefix...
analytics Show cost analytics from the cost tracking database.
availability Show GPU availability / stock status across providers.
budget-optimize Find optimal deployment under budget constraints
checkpoint Manage distributed checkpoints.
cleanup Clean up unused resources and temporary files
configure Configure cloud provider credentials for GPU...
cost-scaler Cost-aware scaling manager for inference optimization
environments Environment management and promotion
eval Model and endpoint evaluation with baseline comparison
execute Execute shell commands on provisioned GPU instances...
export Export current state or job as Argo-compatible YAML...
gateway Launch an API gateway for inference serving.
gitops GitOps automation and infrastructure as code
helm-generate Generate Helm charts from Terradev workloads
hf-space One-click HuggingFace Spaces deployment
hf-spaces HuggingFace Spaces — create, list, manage, delete.
import Import and register Argo-compatible YAML pipeline
infer Deploy and manage inference endpoints
inferx InferX serverless inference platform - <2s cold...
integrations Show status of observability & ML integrations and...
job Run Terradev job from YAML configuration
k8s Kubernetes cluster management with multi-cloud GPU nodes
karpenter Karpenter GPU node auto-provisioning — install,...
lineage Artifact lineage and tracking
local Local GPU discovery and hybrid compute pool management.
lora Production-grade LoRA adapter management with registry...
manage Manage provisioned GPU instances via provider APIs.
manifests List cached manifests, imported pipelines, and recordings
mcp Run Terradev as an MCP server for agent integration.
migrate Cross-provider workload migration with dry-run analysis
ml ML Platform Integration Commands
model-router Model routing cost/quality-aware routing between...
monitor Monitor GPU utilization, training metrics, and cost.
observe Unified observability pipeline - W&B, Phoenix, Cost...
onboarding Run the interactive onboarding flow
optimize Multi-dimensional optimization: cost + performance +...
orchestrator Model orchestrator for multi-model inference
percentiles Show historical price percentiles (p10p99) per provider.
preflight Run preflight hardware validation on GPU nodes.
price-discovery Enhanced price discovery with capacity and confidence...
providers Manage custom provider profiles for intelligent routing
provision Provision GPU instances across multiple clouds with...
quote Get real-time GPU pricing quotes from all configured...
record Record and export live workflows
reliability Show provider reliability scores and error rates.
retrain Drift-triggered continuous fine-tuning.
rollback EXPLICIT ROLLBACK (versioned manifests)
run One-command GPU provisioning, Docker deployment, and...
schedule Spot-aware scheduling for cost-optimized job execution
setup Get step-by-step setup instructions for any cloud...
smart-deploy Smart deployment with automatic optimization
sso Enterprise SSO authentication
stage Compress, chunk, and pre-position datasets near compute.
status Show current status of all provisioned instances and...
train Launch distributed training jobs across provisioned...
triggers Event-driven automation and triggers
up CLI-native provisioning with manifest cache + drift...
warm-pool Warm pool manager for intelligent pre-warming
terradev agent▾
Explainer coming soon — add a paragraph under this heading in CLI.md to replace this placeholder.
Usage: terradev agent [OPTIONS] COMMAND [ARGS]...
Provision and manage heterogeneous agent fleets.
Multi-tier GPU provisioning purpose-built for multi-agent LLM workloads.
Automatically maps agent count to hardware tiers based on empirical workload
research (decode-dominated, KV cache preservation critical).
Tiers provisioned:
reasoning — H100 SXM: long-context KV preservation (P95: 120K tokens)
decode — A100 80GB: memory-bandwidth-optimised token streaming
cpu_tools — 48-vCPU: Bash/WebFetch/file-op tool execution
Examples:
terradev agent plan --agents 16 --model meta-llama/Llama-3.1-70B-Instruct
terradev agent deploy --agents 16 --model meta-llama/Llama-3.1-70B-Instruct
terradev agent deploy --topology ./agent-fleet.yaml
terradev agent status --fleet-id ag_abc123
terradev agent scale --fleet-id ag_abc123 --tier decode --count 8
terradev agent cost --fleet-id ag_abc123
terradev agent list
terradev agent teardown --fleet-id ag_abc123
Options:
--help Show this message and exit.
Commands:
cost Show real-time cost breakdown for a fleet by tier.
deploy Provision a heterogeneous agent fleet across all tiers...
list List all known agent fleets.
plan Plan a heterogeneous agent fleet without provisioning.
scale Scale a single fleet tier up or down without affecting other...
status Show live status of a fleet — tier health, KV hit rate, queue...
teardown Terminate all fleet instances and remove fleet state.
terradev agent cost▾
Explainer coming soon — add a paragraph under this heading in CLI.md to replace this placeholder.
Usage: terradev agent cost [OPTIONS]
Show real-time cost breakdown for a fleet by tier.
Example:
terradev agent cost --fleet-id ag_abc123
Options:
--fleet-id TEXT Fleet ID [required]
--format [table|json]
--help Show this message and exit.
terradev agent deploy▾
Explainer coming soon — add a paragraph under this heading in CLI.md to replace this placeholder.
Usage: terradev agent deploy [OPTIONS]
Provision a heterogeneous agent fleet across all tiers simultaneously.
Provisions reasoning (H100), decode (A100), and CPU tools tiers in parallel
using the existing DAGExecutor wave-parallel orchestration.
Examples:
terradev agent deploy --agents 16 --model meta-llama/Llama-3.1-70B-Instruct
terradev agent deploy --agents 32 --dry-run
terradev agent deploy --topology ./agent-fleet.yaml
terradev agent deploy --agents 8 --planner-gpu H100_SXM --worker-gpu A100_SXM_80
Options:
-n, --agents INTEGER Number of concurrent agent loops
-m, --model TEXT Model to serve
--reasoning [instant|thinking]
--topology PATH Path to agent-fleet.yaml spec file
--planner-gpu TEXT Reasoning tier GPU type
--planner-count INTEGER Reasoning tier instance count
--worker-gpu TEXT Decode tier GPU type
--worker-count INTEGER Decode tier instance count
--cpu-cores INTEGER vCPU count for CPU tools tier
-p, --providers TEXT Cloud providers to use (e.g. runpod vastai)
--max-price FLOAT Max price per GPU/hr in USD
--dry-run Show allocation plan without provisioning
--format [table|json]
--help Show this message and exit.
terradev agent list▾
Explainer coming soon — add a paragraph under this heading in CLI.md to replace this placeholder.
Usage: terradev agent list [OPTIONS]
List all known agent fleets.
Example:
terradev agent list
Options:
--format [table|json]
--help Show this message and exit.
terradev agent plan▾
Explainer coming soon — add a paragraph under this heading in CLI.md to replace this placeholder.
Usage: terradev agent plan [OPTIONS]
Plan a heterogeneous agent fleet without provisioning.
Shows the recommended tier configuration, hardware selection rationale, KV
cache budget, and cost estimate based on arXiv:2605.26297 research.
Examples:
terradev agent plan --agents 16 --model meta-llama/Llama-3.1-70B-Instruct
terradev agent plan --agents 32 --model meta-llama/Llama-3.1-8B-Instruct --format json
terradev agent plan --agents 8 --planner-gpu H100_SXM --worker-gpu A100_SXM_80
Options:
-n, --agents INTEGER Number of concurrent agent loops to
provision for [required]
-m, --model TEXT Model to serve across the fleet
--reasoning [instant|thinking] Reasoning mode: instant (faster) or thinking
(extended CoT, 45-67% more output tokens)
--planner-gpu TEXT Override reasoning tier GPU type (e.g.
H100_SXM)
--planner-count INTEGER Override reasoning tier instance count
--worker-gpu TEXT Override decode tier GPU type (e.g.
A100_SXM_80)
--worker-count INTEGER Override decode tier instance count
--cpu-cores INTEGER vCPU count for CPU tools tier instances
--format [table|json] Output format
--help Show this message and exit.
terradev agent scale▾
Explainer coming soon — add a paragraph under this heading in CLI.md to replace this placeholder.
Usage: terradev agent scale [OPTIONS]
Scale a single fleet tier up or down without affecting other tiers.
KV cache state on existing instances is PRESERVED during scale operations.
New instances are added to the pool; the router distributes new requests to
them.
Examples:
terradev agent scale --fleet-id ag_abc123 --tier decode --count 8
terradev agent scale --fleet-id ag_abc123 --tier reasoning --count 3
terradev agent scale --fleet-id ag_abc123 --tier cpu_tools --count 4
Options:
--fleet-id TEXT Fleet ID [required]
--tier [reasoning|decode|cpu_tools]
Tier to scale [required]
--count INTEGER New instance count for this tier [required]
-p, --providers TEXT Providers to use for scale-out instances
--help Show this message and exit.
terradev agent status▾
Explainer coming soon — add a paragraph under this heading in CLI.md to replace this placeholder.
Usage: terradev agent status [OPTIONS]
Show live status of a fleet — tier health, KV hit rate, queue depth, cost.
Key metrics explained: kv_hit_rate — target >0.85. Below 0.80 = cache
thrashing (expensive recompute). ttft_p95_ms — reasoning tier target
<2000ms. Above = scale out reasoning. queue_depth — decode tier pending
requests. Above 6 = scale out decode.
(Metrics from arXiv:2605.26297 empirical benchmarking)
Options:
--fleet-id TEXT Fleet ID returned by 'terradev agent deploy'
[required]
--format [table|json]
--help Show this message and exit.
terradev agent teardown▾
Explainer coming soon — add a paragraph under this heading in CLI.md to replace this placeholder.
Usage: terradev agent teardown [OPTIONS]
Terminate all fleet instances and remove fleet state.
Example:
terradev agent teardown --fleet-id ag_abc123
terradev agent teardown --fleet-id ag_abc123 --yes
Options:
--fleet-id TEXT Fleet ID to destroy [required]
--yes Skip confirmation prompt
--help Show this message and exit.
terradev agentic-serving▾
Explainer coming soon — add a paragraph under this heading in CLI.md to replace this placeholder.
Usage: terradev agentic-serving [OPTIONS] COMMAND [ARGS]...
Agentic inference serving KV cache TTL, prefix caching, LMCache, priority
scheduling.
Options:
--help Show this message and exit.
Commands:
configure Configure agentic inference serving settings.
helm-values Print Helm values for agentic inference deployment.
k8s Print K8s deployment manifests for agentic inference.
launch-args Print engine launch arguments for copy-paste.
lmcache-env Print LMCache environment variables.
show-config Show current agentic serving configuration.
terradev agentic-serving configure▾
Explainer coming soon — add a paragraph under this heading in CLI.md to replace this placeholder.
Usage: terradev agentic-serving configure [OPTIONS]
Configure agentic inference serving settings.
Options:
--engine [vllm|sglang] Inference engine
--model TEXT
--tp INTEGER Tensor parallel size
--max-model-len INTEGER
--gpu-mem FLOAT
--lmcache / --no-lmcache Enable LMCache KV offload
--lmcache-backend [cpu|disk|redis]
--disaggregation / --no-disaggregation
Prefill-decode disaggregation
--help Show this message and exit.
terradev agentic-serving helm-values▾
Explainer coming soon — add a paragraph under this heading in CLI.md to replace this placeholder.
Usage: terradev agentic-serving helm-values [OPTIONS]
Print Helm values for agentic inference deployment.
Options:
-f, --format [json|yaml]
--help Show this message and exit.
terradev agentic-serving k8s▾
Explainer coming soon — add a paragraph under this heading in CLI.md to replace this placeholder.
Usage: terradev agentic-serving k8s [OPTIONS]
Print K8s deployment manifests for agentic inference.
Options:
-n, --namespace TEXT K8s namespace
--help Show this message and exit.
terradev agentic-serving launch-args▾
Explainer coming soon — add a paragraph under this heading in CLI.md to replace this placeholder.
Usage: terradev agentic-serving launch-args [OPTIONS]
Print engine launch arguments for copy-paste.
Options:
--help Show this message and exit.
terradev agentic-serving lmcache-env▾
Explainer coming soon — add a paragraph under this heading in CLI.md to replace this placeholder.
Usage: terradev agentic-serving lmcache-env [OPTIONS]
Print LMCache environment variables.
Options:
--help Show this message and exit.
terradev agentic-serving show-config▾
Explainer coming soon — add a paragraph under this heading in CLI.md to replace this placeholder.
Usage: terradev agentic-serving show-config [OPTIONS]
Show current agentic serving configuration.
Options:
-f, --format [json|text]
--help Show this message and exit.
terradev analytics▾
Explainer coming soon — add a paragraph under this heading in CLI.md to replace this placeholder.
Usage: terradev analytics [OPTIONS]
Show cost analytics from the cost tracking database.
Options:
-d, --days INTEGER Number of days to analyze (default: 7)
-f, --format [table|json] Output format
--help Show this message and exit.
terradev availability▾
Explainer coming soon — add a paragraph under this heading in CLI.md to replace this placeholder.
Usage: terradev availability [OPTIONS]
Show GPU availability / stock status across providers.
Options:
-g, --gpu-type TEXT GPU type filter (shows all if omitted)
-w, --window INTEGER Lookback window in hours (default: 24)
--help Show this message and exit.
terradev budget-optimize▾
Explainer coming soon — add a paragraph under this heading in CLI.md to replace this placeholder.
Usage: terradev budget-optimize [OPTIONS]
Find optimal deployment under budget constraints
Options:
--gpu-type TEXT GPU type [required]
--budget FLOAT Budget constraint ($/hr) [required]
--gpu-count INTEGER Number of GPUs
--hours FLOAT Estimated runtime in hours
--region TEXT Preferred region
--workload TEXT Workload type
--help Show this message and exit.
terradev checkpoint▾
Explainer coming soon — add a paragraph under this heading in CLI.md to replace this placeholder.
Usage: terradev checkpoint [OPTIONS] {list|restore|promote|delete}
Manage distributed checkpoints.
Local filesystem by default. Supports manifest-based atomic commits,
parallel shard verification, and retention policies.
Examples: terradev checkpoint list -j job-abc123 terradev checkpoint
restore -j job-abc123 terradev checkpoint restore -j job-abc123 --step
5000 terradev checkpoint promote -j job-abc123 --checkpoint-id ckpt-xyz
--dest /models/final terradev checkpoint delete -j job-abc123
--checkpoint-id ckpt-xyz
Options:
-j, --job-id TEXT Job ID [required]
--step INTEGER Checkpoint step
--checkpoint-id TEXT Checkpoint ID
--dest TEXT Destination path (for promote)
-f, --format [json|text]
--help Show this message and exit.
terradev cleanup▾
Explainer coming soon — add a paragraph under this heading in CLI.md to replace this placeholder.
Usage: terradev cleanup [OPTIONS]
Clean up unused resources and temporary files
Options:
--help Show this message and exit.
terradev configure▾
Explainer coming soon — add a paragraph under this heading in CLI.md to replace this placeholder.
Usage: terradev configure [OPTIONS]
Configure cloud provider credentials for GPU provisioning.
Stores API keys locally at ~/.terradev/credentials.json (never sent to
Terradev servers).
Examples: terradev configure --provider runpod terradev configure
--provider aws terradev configure # Interactive mode for all
providers
Quick Start: RunPod is the easiest to set up (5 minutes): terradev setup
runpod --quick
Options:
-p, --provider TEXT Configure specific provider (e.g., runpod, vastai, aws)
--help Show this message and exit.
terradev cost-scaler▾
Explainer coming soon — add a paragraph under this heading in CLI.md to replace this placeholder.
Usage: terradev cost-scaler [OPTIONS] COMMAND [ARGS]...
Cost-aware scaling manager for inference optimization
Options:
--help Show this message and exit.
Commands:
model-details Get cost details for a specific model
start Start the cost-aware scaling manager
status Get cost scaler status and recommendations
terradev cost-scaler model-details▾
Explainer coming soon — add a paragraph under this heading in CLI.md to replace this placeholder.
Usage: terradev cost-scaler model-details [OPTIONS] MODEL_ID
Get cost details for a specific model
Options:
--help Show this message and exit.
terradev cost-scaler start▾
Explainer coming soon — add a paragraph under this heading in CLI.md to replace this placeholder.
Usage: terradev cost-scaler start [OPTIONS]
Start the cost-aware scaling manager
Options:
--strategy [minimize_cost|balance_cost_latency|latency_critical|budget_constrained]
Cost optimization strategy
--budget FLOAT Hourly budget in USD
--cost-per-gb FLOAT Cost per GB per hour in USD
--help Show this message and exit.
terradev cost-scaler status▾
Explainer coming soon — add a paragraph under this heading in CLI.md to replace this placeholder.
Usage: terradev cost-scaler status [OPTIONS]
Get cost scaler status and recommendations
Options:
--help Show this message and exit.
terradev environments▾
Explainer coming soon — add a paragraph under this heading in CLI.md to replace this placeholder.
Usage: terradev environments [OPTIONS] COMMAND [ARGS]...
Environment management and promotion
Options:
--help Show this message and exit.
Commands:
approve Approve and execute promotion
history Show promotion history
list List artifacts by environment
promote Request environment promotion
terradev environments approve▾
Explainer coming soon — add a paragraph under this heading in CLI.md to replace this placeholder.
Usage: terradev environments approve [OPTIONS] PROMOTION_ID
Approve and execute promotion
Options:
--user TEXT User approving promotion
--help Show this message and exit.
terradev environments history▾
Explainer coming soon — add a paragraph under this heading in CLI.md to replace this placeholder.
Usage: terradev environments history [OPTIONS]
Show promotion history
Options:
--artifact TEXT Filter by artifact name
--help Show this message and exit.
terradev environments list▾
Explainer coming soon — add a paragraph under this heading in CLI.md to replace this placeholder.
Usage: terradev environments list [OPTIONS]
List artifacts by environment
Options:
--env TEXT Filter by environment
--help Show this message and exit.
terradev environments promote▾
Explainer coming soon — add a paragraph under this heading in CLI.md to replace this placeholder.
Usage: terradev environments promote [OPTIONS] ARTIFACT_NAME
Request environment promotion
Options:
--from [dev|staging|prod] [required]
--to [dev|staging|prod] [required]
--user TEXT User requesting promotion
--help Show this message and exit.
terradev eval▾
Explainer coming soon — add a paragraph under this heading in CLI.md to replace this placeholder.
Usage: terradev eval [OPTIONS] COMMAND [ARGS]...
Model and endpoint evaluation with baseline comparison
Options:
--help Show this message and exit.
Commands:
compare Compare two models side-by-side
evaluation Run model or endpoint evaluation
terradev eval compare▾
Explainer coming soon — add a paragraph under this heading in CLI.md to replace this placeholder.
Usage: terradev eval compare [OPTIONS] MODEL_A MODEL_B
Compare two models side-by-side
Options:
--dataset TEXT Dataset for comparison [required]
--metrics TEXT Metrics to compare
--output TEXT Output file for comparison results
--help Show this message and exit.
terradev eval evaluation▾
Explainer coming soon — add a paragraph under this heading in CLI.md to replace this placeholder.
Usage: terradev eval evaluation [OPTIONS]
Run model or endpoint evaluation
Options:
--model TEXT Model checkpoint path
--endpoint TEXT API endpoint URL
--dataset TEXT Dataset path for evaluation
--metrics TEXT Metrics to evaluate
--baseline TEXT Baseline result file for comparison
--workload-type TEXT Workload type classification
--duration INTEGER Duration for endpoint evaluation (seconds)
--output TEXT Output file for results
--format [json|table]
--help Show this message and exit.
terradev execute▾
Explainer coming soon — add a paragraph under this heading in CLI.md to replace this placeholder.
Usage: terradev execute [OPTIONS]
Execute shell commands on provisioned GPU instances via provider APIs.
Run commands directly on your GPU instances without needing to SSH. Commands
are executed through the provider's API and output is returned.
Examples: terradev execute -i <instance-id> -c "nvidia-smi"
# Check GPU status terradev execute -i <instance-id> -c "python train.py"
# Run training script terradev execute -i <instance-id> -c "ls -la
/workspace" # List files terradev execute -i <instance-id> -c
"pip install torch" --async # Install packages async
Use Cases: - Check GPU utilization: nvidia-smi - Run training scripts:
python train.py --args - Install dependencies: pip install <package> -
Monitor jobs: ps aux | grep python
Instance IDs: Get instance IDs from: terradev status Use the full ID
shown in the status output
Async Mode: Use --async-exec for long-running commands Command runs in
background, returns immediately with job ID
Options:
-i, --instance-id TEXT Instance ID (from terradev status) [required]
--cmd TEXT Command to execute on the instance [required]
--async-exec Run command asynchronously (returns immediately)
--help Show this message and exit.
terradev export▾
Explainer coming soon — add a paragraph under this heading in CLI.md to replace this placeholder.
Usage: terradev export [OPTIONS]
Export current state or job as Argo-compatible YAML pipeline
Options:
-o, --output TEXT Output YAML file path [required]
-j, --job TEXT Specific job to export (omits latest)
--cache-dir TEXT Manifest cache directory
--format [argo|native] Output format (argo-compatible or terradev-native)
--help Show this message and exit.
terradev gateway▾
Explainer coming soon — add a paragraph under this heading in CLI.md to replace this placeholder.
Usage: terradev gateway [OPTIONS]
Launch an API gateway for inference serving.
Provides OpenAI/Anthropic/custom API entry and exit points for inference
workflows. Integrates with Terradev's inference routing and KV cache
management.
OpenAI-compatible endpoints:
- POST /v1/chat/completions
- POST /v1/completions
Anthropic-compatible endpoints:
- POST /v1/messages
- POST /v1/messages/batches
Custom workflow endpoints:
- POST /v1/custom/entry/{workflow_id}
- POST /v1/custom/exit/{workflow_id}
Management endpoints:
- GET /health
- GET /v1/gateway/status
Examples:
terradev gateway
terradev gateway --host 0.0.0.0 --port 8080
terradev gateway --no-anthropic --max-concurrent 50
terradev gateway --model meta-llama/Llama-3.1-8B-Instruct
Testing the gateway:
curl http://localhost:8000/health
curl http://localhost:8000/v1/gateway/status
Options:
-h, --host TEXT Host to bind the gateway server
-p, --port INTEGER Port for the gateway server
--openai Enable OpenAI-compatible endpoints
--no-openai Disable OpenAI-compatible endpoints
--anthropic Enable Anthropic-compatible endpoints
--no-anthropic Disable Anthropic-compatible endpoints
--custom Enable custom workflow endpoints
--no-custom Disable custom workflow endpoints
--max-concurrent INTEGER Maximum concurrent requests
--timeout INTEGER Request timeout in seconds
--cors Enable CORS
--no-cors Disable CORS
--cors-origins TEXT CORS allowed origins
--model TEXT Default model for inference
--no-inference-router Disable inference router integration
--help Show this message and exit.
terradev gitops▾
Explainer coming soon — add a paragraph under this heading in CLI.md to replace this placeholder.
Usage: terradev gitops [OPTIONS] COMMAND [ARGS]...
GitOps automation and infrastructure as code
Options:
--help Show this message and exit.
Commands:
bootstrap Bootstrap GitOps tool on the cluster
init Initialize GitOps repository and structure
sync Sync cluster with Git repository
validate Validate GitOps configuration
terradev gitops bootstrap▾
Explainer coming soon — add a paragraph under this heading in CLI.md to replace this placeholder.
Usage: terradev gitops bootstrap [OPTIONS]
Bootstrap GitOps tool on the cluster
Options:
--tool [argocd|flux] GitOps tool [required]
--cluster TEXT Cluster name [required]
--namespace TEXT Namespace for GitOps tools
--help Show this message and exit.
terradev gitops init▾
Explainer coming soon — add a paragraph under this heading in CLI.md to replace this placeholder.
Usage: terradev gitops init [OPTIONS]
Initialize GitOps repository and structure
Options:
--provider [github|gitlab|bitbucket|azure_devops]
Git provider [required]
--repo, --repository TEXT Repository name (format: owner/repo)
[required]
--tool [argocd|flux] GitOps tool
--cluster TEXT Cluster name [required]
--git-url TEXT Git repository URL (auto-generated if not
provided)
--git-token TEXT Git access token
--namespace TEXT Namespace for GitOps tools
--auto-sync / --no-auto-sync Enable automatic synchronization
--prune / --no-prune Enable resource pruning
--help Show this message and exit.
terradev gitops sync▾
Explainer coming soon — add a paragraph under this heading in CLI.md to replace this placeholder.
Usage: terradev gitops sync [OPTIONS]
Sync cluster with Git repository
Options:
--cluster TEXT Cluster name [required]
--environment TEXT Environment to sync
--tool [argocd|flux] GitOps tool
--help Show this message and exit.
terradev gitops validate▾
Explainer coming soon — add a paragraph under this heading in CLI.md to replace this placeholder.
Usage: terradev gitops validate [OPTIONS]
Validate GitOps configuration
Options:
--dry-run / --apply Dry run validation or apply changes
--cluster TEXT Cluster name for validation
--environment TEXT Environment to validate
--help Show this message and exit.
terradev helm-generate▾
Explainer coming soon — add a paragraph under this heading in CLI.md to replace this placeholder.
Usage: terradev helm-generate [OPTIONS]
Generate Helm charts from Terradev workloads
Options:
--workload TEXT Workload type (training, inference, cost-optimized,
high-performance, moe-inference, rag, vllm-optimized)
--gpu-type TEXT GPU type (A100, H100, V100, L4, L40S, RTX 4090, T4,
etc.)
--image TEXT Docker image [required]
--gpu-count INTEGER Number of GPUs
--memory INTEGER Memory in GB
--storage INTEGER Storage in GB
--budget FLOAT Budget constraint ($/hr)
--region TEXT Preferred region
--port INTEGER Expose port(s) via Service (repeatable)
-s, --stack TEXT Stack integrations: qdrant, phoenix, guardrails
(repeatable)
-o, --output TEXT Output directory
--name TEXT Chart name
--dry-run Show chart config without generating
--help Show this message and exit.
terradev hf-space▾
Explainer coming soon — add a paragraph under this heading in CLI.md to replace this placeholder.
Usage: terradev hf-space [OPTIONS] SPACE_NAME
One-click HuggingFace Spaces deployment
Options:
--model-id TEXT HuggingFace model ID to deploy [required]
--hardware [cpu-basic|cpu-upgrade|t4-medium|a10g-large|a100-large]
Hardware tier for the Space
--sdk [gradio|streamlit|docker]
SDK for the Space
--private Make the Space private
--template [llm|embedding|image]
Use pre-configured template
-e, --env TEXT Environment variables KEY=VALUE
-s, --secret TEXT Secrets KEY=VALUE
--help Show this message and exit.
terradev hf-spaces▾
Explainer coming soon — add a paragraph under this heading in CLI.md to replace this placeholder.
Usage: terradev hf-spaces [OPTIONS] COMMAND [ARGS]...
HuggingFace Spaces — create, list, manage, delete.
Options:
--help Show this message and exit.
Commands:
create Create a new Space with auto-generated app.
delete Delete a Space.
hardware Show or change hardware tier.
info Get Space details.
list List HuggingFace Spaces.
logs Show Space build/run logs.
pause Pause a running Space (stops billing).
restart Restart a Space (factory reboot).
resume Resume a paused Space.
terradev hf-spaces create▾
Explainer coming soon — add a paragraph under this heading in CLI.md to replace this placeholder.
Usage: terradev hf-spaces create [OPTIONS] SPACE_NAME
Create a new Space with auto-generated app.
Options:
--model-id TEXT [required]
--hardware [cpu-basic|cpu-upgrade|t4-medium|a10g-large|a100-large]
--sdk [gradio|streamlit|docker]
--private
--template [llm|embedding|image]
--help Show this message and exit.
terradev hf-spaces delete▾
Explainer coming soon — add a paragraph under this heading in CLI.md to replace this placeholder.
Usage: terradev hf-spaces delete [OPTIONS] SPACE_ID
Delete a Space.
Options:
-y, --yes
--help Show this message and exit.
terradev hf-spaces hardware▾
Explainer coming soon — add a paragraph under this heading in CLI.md to replace this placeholder.
Usage: terradev hf-spaces hardware [OPTIONS] SPACE_ID
Show or change hardware tier.
Options:
--set [cpu-basic|cpu-upgrade|t4-medium|a10g-large|a100-large]
--help Show this message and exit.
terradev hf-spaces info▾
Explainer coming soon — add a paragraph under this heading in CLI.md to replace this placeholder.
Usage: terradev hf-spaces info [OPTIONS] SPACE_ID
Get Space details.
Options:
-f, --format [json|text]
--help Show this message and exit.
terradev hf-spaces list▾
Explainer coming soon — add a paragraph under this heading in CLI.md to replace this placeholder.
Usage: terradev hf-spaces list [OPTIONS]
List HuggingFace Spaces.
Options:
--author TEXT
-n, --limit INTEGER
-f, --format [json|text]
--help Show this message and exit.
terradev hf-spaces logs▾
Explainer coming soon — add a paragraph under this heading in CLI.md to replace this placeholder.
Usage: terradev hf-spaces logs [OPTIONS] SPACE_ID
Show Space build/run logs.
Options:
--help Show this message and exit.
terradev hf-spaces pause▾
Explainer coming soon — add a paragraph under this heading in CLI.md to replace this placeholder.
Usage: terradev hf-spaces pause [OPTIONS] SPACE_ID
Pause a running Space (stops billing).
Options:
--help Show this message and exit.
terradev hf-spaces restart▾
Explainer coming soon — add a paragraph under this heading in CLI.md to replace this placeholder.
Usage: terradev hf-spaces restart [OPTIONS] SPACE_ID
Restart a Space (factory reboot).
Options:
--help Show this message and exit.
terradev hf-spaces resume▾
Explainer coming soon — add a paragraph under this heading in CLI.md to replace this placeholder.
Usage: terradev hf-spaces resume [OPTIONS] SPACE_ID
Resume a paused Space.
Options:
--help Show this message and exit.
terradev import▾
Explainer coming soon — add a paragraph under this heading in CLI.md to replace this placeholder.
Usage: terradev import [OPTIONS] YAML_FILE
Import and register Argo-compatible YAML pipeline
Options:
-n, --name TEXT Name to register pipeline (defaults to YAML metadata name)
--force Overwrite existing pipeline with same name
--validate-only Only validate, do not register
--cache-dir TEXT Manifest cache directory
--help Show this message and exit.
terradev infer▾
Explainer coming soon — add a paragraph under this heading in CLI.md to replace this placeholder.
Usage: terradev infer [OPTIONS] COMMAND [ARGS]...
Deploy and manage inference endpoints
Options:
--help Show this message and exit.
Commands:
deploy Deploy and manage inference endpoints
endpoint Deploy inference endpoint
failover Run health checks and auto-failover for inference endpoints.
route Find the best inference endpoint using latency-aware routing.
status Show inference endpoint health, latency, and failover status.
terradev infer deploy▾
Explainer coming soon — add a paragraph under this heading in CLI.md to replace this placeholder.
Usage: terradev infer deploy [OPTIONS]
Deploy and manage inference endpoints
Options:
-m, --model TEXT Model name or path [required]
-t, --type [llm|embedding|vision]
Model type
-p, --provider [runpod|vastai|lambda_labs|baseten]
Provider preference
-g, --gpu-type TEXT GPU type preference
-r, --region TEXT Region preference
--max-latency FLOAT Max latency in ms
--max-cost FLOAT Max cost per request
--help Show this message and exit.
terradev infer endpoint▾
Explainer coming soon — add a paragraph under this heading in CLI.md to replace this placeholder.
Usage: terradev infer endpoint [OPTIONS] MODEL_PATH
Deploy inference endpoint
Options:
-n, --name TEXT Endpoint name (required) [required]
-p, --provider [runpod|vastai|lambda_labs|baseten]
Provider (runpod|vastai|lambda_labs|baseten)
-g, --gpu-type TEXT GPU type (A100|H100|RTX4090)
--min-workers INTEGER Minimum workers
--max-workers INTEGER Maximum workers
--idle-timeout INTEGER Idle timeout in seconds
--cost-optimize Enable cost optimization
--dry-run Show deployment plan without deploying
--help Show this message and exit.
terradev infer failover▾
Explainer coming soon — add a paragraph under this heading in CLI.md to replace this placeholder.
Usage: terradev infer failover [OPTIONS]
Run health checks and auto-failover for inference endpoints.
Probes all registered inference endpoints. If a primary endpoint is
unhealthy and has a backup configured, traffic automatically shifts to the
backup provider.
Open source feature - available to all users.
Options:
--dry-run Show what would happen without executing failover
--help Show this message and exit.
terradev infer route▾
Explainer coming soon — add a paragraph under this heading in CLI.md to replace this placeholder.
Usage: terradev infer route [OPTIONS]
Find the best inference endpoint using latency-aware routing.
Selects the optimal healthy endpoint based on strategy: - latency: lowest
average response time (default) - cost: cheapest price per hour - score:
weighted combination of latency + cost
Use --measure to run fresh ping/TTFB probes before selecting.
Integrates with WebPageTest TTFB probes for real-world latency data.
Set WPT_API_KEY env var to enable WebPageTest integration.
Options:
-m, --model TEXT Filter by model name
-s, --strategy [latency|cost|score]
Routing strategy (default: latency)
--measure Run fresh latency measurements before
routing
--help Show this message and exit.
terradev infer status▾
Explainer coming soon — add a paragraph under this heading in CLI.md to replace this placeholder.
Usage: terradev infer status [OPTIONS]
Show inference endpoint health, latency, and failover status.
Displays all registered inference endpoints with their health state, average
latency, provider, and failover configuration.
Use --check to run live health probes before displaying.
Options:
--check Run live health probes before showing status
--help Show this message and exit.
terradev inferx▾
Explainer coming soon — add a paragraph under this heading in CLI.md to replace this placeholder.
Usage: terradev inferx [OPTIONS] COMMAND [ARGS]...
InferX serverless inference platform - <2s cold starts, 90% GPU utilization
Options:
--help Show this message and exit.
Commands:
deploy Deploy model to InferX serverless platform
inferx-configure Configure InferX provider credentials
inferx-delete Delete model deployment
inferx-optimize Analyze and optimize InferX costs with AI-powered...
inferx-quote Get pricing quotes for InferX
inferx-status Get model deployment status
list List all deployed models
usage Get account usage statistics
terradev inferx deploy▾
Explainer coming soon — add a paragraph under this heading in CLI.md to replace this placeholder.
Usage: terradev inferx deploy [OPTIONS]
Deploy model to InferX serverless platform
Options:
--model TEXT Model ID or HuggingFace model name
[required]
--image TEXT Docker image for model
--gpu-type TEXT GPU type
--gpu-memory INTEGER GPU memory in GB
--max-concurrency INTEGER Maximum concurrent requests
--framework TEXT Model framework
--openai-compatible / --no-openai-compatible
OpenAI-compatible API
--timeout INTEGER Request timeout in seconds
--help Show this message and exit.
terradev inferx inferx-configure▾
Explainer coming soon — add a paragraph under this heading in CLI.md to replace this placeholder.
Usage: terradev inferx inferx-configure [OPTIONS]
Configure InferX provider credentials
Options:
--api-key TEXT InferX API key [required]
--endpoint TEXT InferX API endpoint
--region TEXT Region for deployment
--snapshot / --no-snapshot Enable snapshot technology
--gpu-slicing / --no-gpu-slicing
Enable GPU slicing
--multi-tenant / --no-multi-tenant
Enable multi-tenant isolation
--help Show this message and exit.
terradev inferx inferx-delete▾
Explainer coming soon — add a paragraph under this heading in CLI.md to replace this placeholder.
Usage: terradev inferx inferx-delete [OPTIONS]
Delete model deployment
Options:
--model-id TEXT Model deployment ID [required]
--help Show this message and exit.
terradev inferx inferx-optimize▾
Explainer coming soon — add a paragraph under this heading in CLI.md to replace this placeholder.
Usage: terradev inferx inferx-optimize [OPTIONS]
Analyze and optimize InferX costs with AI-powered recommendations
Options:
--cluster-config TEXT Cluster configuration file
--usage-metrics TEXT Usage metrics file
--tier [economy|balanced|performance]
Cost optimization tier
--output TEXT Output file for cost report
--implement Implement cost optimizations automatically
--help Show this message and exit.
terradev inferx inferx-quote▾
Explainer coming soon — add a paragraph under this heading in CLI.md to replace this placeholder.
Usage: terradev inferx inferx-quote [OPTIONS]
Get pricing quotes for InferX
Options:
--gpu-type TEXT GPU type to quote
--region TEXT Region for quote
--help Show this message and exit.
terradev inferx inferx-status▾
Explainer coming soon — add a paragraph under this heading in CLI.md to replace this placeholder.
Usage: terradev inferx inferx-status [OPTIONS]
Get model deployment status
Options:
--model-id TEXT Model deployment ID [required]
--help Show this message and exit.
terradev inferx list▾
Explainer coming soon — add a paragraph under this heading in CLI.md to replace this placeholder.
Usage: terradev inferx list [OPTIONS]
List all deployed models
Options:
--help Show this message and exit.
terradev inferx usage▾
Explainer coming soon — add a paragraph under this heading in CLI.md to replace this placeholder.
Usage: terradev inferx usage [OPTIONS]
Get account usage statistics
Options:
--help Show this message and exit.
terradev integrations▾
Explainer coming soon — add a paragraph under this heading in CLI.md to replace this placeholder.
Usage: terradev integrations [OPTIONS]
Show status of observability & ML integrations and export configs.
Terradev facilitates connections to your existing tools your keys stay
local, and all data flows directly from your instances to your services.
Examples: terradev integrations terradev integrations --export-
grafana terradev integrations --export-scrape-config terradev
integrations --export-wandb-script
Options:
--export-grafana Export a Grafana dashboard JSON for Terradev metrics
--export-scrape-config Print a Prometheus scrape config snippet
--export-wandb-script Print a W&B setup script for remote instances
--help Show this message and exit.
terradev job▾
Explainer coming soon — add a paragraph under this heading in CLI.md to replace this placeholder.
Usage: terradev job [OPTIONS] JOB_FILE
Run Terradev job from YAML configuration
Options:
--optimize TEXT Optimization criteria (cost, latency, balanced)
--help Show this message and exit.
terradev k8s▾
Explainer coming soon — add a paragraph under this heading in CLI.md to replace this placeholder.
Usage: terradev k8s [OPTIONS] COMMAND [ARGS]...
Kubernetes cluster management with multi-cloud GPU nodes
Options:
--help Show this message and exit.
Commands:
create Create multi-cloud Kubernetes GPU cluster
destroy Destroy Kubernetes cluster
info Get detailed cluster information
list List all Kubernetes clusters
terradev k8s create▾
Explainer coming soon — add a paragraph under this heading in CLI.md to replace this placeholder.
Usage: terradev k8s create [OPTIONS] CLUSTER_NAME
Create multi-cloud Kubernetes GPU cluster
Options:
-g, --gpu TEXT GPU type (H100, A100, L40) [required]
-n, --count INTEGER Number of GPU nodes [required]
--max-price FLOAT Maximum price per hour
--multi-cloud Use multi-cloud provisioning
--prefer-spot Prefer spot instances
--aws-region TEXT AWS region
--gcp-region TEXT GCP region
--control-plane [eks|gke|self-hosted]
Control plane type
--help Show this message and exit.
terradev k8s destroy▾
Explainer coming soon — add a paragraph under this heading in CLI.md to replace this placeholder.
Usage: terradev k8s destroy [OPTIONS] CLUSTER_NAME
Destroy Kubernetes cluster
Options:
--help Show this message and exit.
terradev k8s info▾
Explainer coming soon — add a paragraph under this heading in CLI.md to replace this placeholder.
Usage: terradev k8s info [OPTIONS] CLUSTER_NAME
Get detailed cluster information
Options:
--help Show this message and exit.
terradev k8s list▾
Explainer coming soon — add a paragraph under this heading in CLI.md to replace this placeholder.
Usage: terradev k8s list [OPTIONS]
List all Kubernetes clusters
Options:
--help Show this message and exit.
terradev karpenter▾
Explainer coming soon — add a paragraph under this heading in CLI.md to replace this placeholder.
Usage: terradev karpenter [OPTIONS] COMMAND [ARGS]...
Karpenter GPU node auto-provisioning — install, nodepools, status, events.
Options:
--help Show this message and exit.
Commands:
create-nodepool Create topology-optimized NodePool + EC2NodeClass.
delete-nodepool Delete a NodePool and its EC2NodeClass.
events Show recent Karpenter provisioning events.
gpu-nodes List GPU nodes with capacity details.
install Install or upgrade Karpenter via Helm.
logs Tail Karpenter controller logs.
nodepools List all Karpenter NodePools.
resources Show cluster CPU, memory, GPU usage.
status Show Karpenter health, node pools, GPU summary.
terradev karpenter create-nodepool▾
Explainer coming soon — add a paragraph under this heading in CLI.md to replace this placeholder.
Usage: terradev karpenter create-nodepool [OPTIONS]
Create topology-optimized NodePool + EC2NodeClass.
Options:
-g, --gpu-type TEXT H100, A100, A10G, L40S, L4, T4, V100 [required]
--cpu-limit TEXT
--memory-limit TEXT
--help Show this message and exit.
terradev karpenter delete-nodepool▾
Explainer coming soon — add a paragraph under this heading in CLI.md to replace this placeholder.
Usage: terradev karpenter delete-nodepool [OPTIONS] NAME
Delete a NodePool and its EC2NodeClass.
Options:
-y, --yes
--help Show this message and exit.
terradev karpenter events▾
Explainer coming soon — add a paragraph under this heading in CLI.md to replace this placeholder.
Usage: terradev karpenter events [OPTIONS]
Show recent Karpenter provisioning events.
Options:
-n, --limit INTEGER
--help Show this message and exit.
terradev karpenter gpu-nodes▾
Explainer coming soon — add a paragraph under this heading in CLI.md to replace this placeholder.
Usage: terradev karpenter gpu-nodes [OPTIONS]
List GPU nodes with capacity details.
Options:
-f, --format [json|text]
--help Show this message and exit.
terradev karpenter install▾
Explainer coming soon — add a paragraph under this heading in CLI.md to replace this placeholder.
Usage: terradev karpenter install [OPTIONS]
Install or upgrade Karpenter via Helm.
Options:
-v, --version TEXT
--cluster-name TEXT [required]
--help Show this message and exit.
terradev karpenter logs▾
Explainer coming soon — add a paragraph under this heading in CLI.md to replace this placeholder.
Usage: terradev karpenter logs [OPTIONS]
Tail Karpenter controller logs.
Options:
-n, --lines INTEGER
--help Show this message and exit.
terradev karpenter nodepools▾
Explainer coming soon — add a paragraph under this heading in CLI.md to replace this placeholder.
Usage: terradev karpenter nodepools [OPTIONS]
List all Karpenter NodePools.
Options:
-f, --format [json|text]
--help Show this message and exit.
terradev karpenter resources▾
Explainer coming soon — add a paragraph under this heading in CLI.md to replace this placeholder.
Usage: terradev karpenter resources [OPTIONS]
Show cluster CPU, memory, GPU usage.
Options:
-f, --format [json|text]
--help Show this message and exit.
terradev karpenter status▾
Explainer coming soon — add a paragraph under this heading in CLI.md to replace this placeholder.
Usage: terradev karpenter status [OPTIONS]
Show Karpenter health, node pools, GPU summary.
Options:
-f, --format [json|text]
--help Show this message and exit.
terradev lineage▾
Explainer coming soon — add a paragraph under this heading in CLI.md to replace this placeholder.
Usage: terradev lineage [OPTIONS] COMMAND [ARGS]...
Artifact lineage and tracking
Options:
--help Show this message and exit.
Commands:
add-input Add input artifact to execution (manual override)
add-output Add output artifact to execution (manual override)
auto Start automatic lineage tracking for a pipeline execution
complete Complete execution and finalize lineage record
diff Compare two pipeline executions
export Export lineage data for compliance reports
graph Show lineage graph for artifact
production Show artifacts in production environment
register Register a new artifact for lineage tracking
show Show complete provenance of a model (auto-generated)
trace Trace complete lineage from checkpoint or execution
terradev lineage add-input▾
Explainer coming soon — add a paragraph under this heading in CLI.md to replace this placeholder.
Usage: terradev lineage add-input [OPTIONS] EXECUTION_ID
{dataset|model|config|checkpoint}
ARTIFACT_ID
Add input artifact to execution (manual override)
Options:
--help Show this message and exit.
terradev lineage add-output▾
Explainer coming soon — add a paragraph under this heading in CLI.md to replace this placeholder.
Usage: terradev lineage add-output [OPTIONS] EXECUTION_ID
{model|checkpoint|metrics|evaluation}
ARTIFACT_ID
Add output artifact to execution (manual override)
Options:
--help Show this message and exit.
terradev lineage auto▾
Explainer coming soon — add a paragraph under this heading in CLI.md to replace this placeholder.
Usage: terradev lineage auto [OPTIONS]
Start automatic lineage tracking for a pipeline execution
Options:
--pipeline TEXT Pipeline ID [required]
--env [dev|staging|prod] Execution environment
--triggered-by TEXT Who triggered this execution
--help Show this message and exit.
terradev lineage complete▾
Explainer coming soon — add a paragraph under this heading in CLI.md to replace this placeholder.
Usage: terradev lineage complete [OPTIONS] EXECUTION_ID
Complete execution and finalize lineage record
Options:
--status TEXT Final status (completed, failed)
--help Show this message and exit.
terradev lineage diff▾
Explainer coming soon — add a paragraph under this heading in CLI.md to replace this placeholder.
Usage: terradev lineage diff [OPTIONS] VERSION1 VERSION2
Compare two pipeline executions
Options:
--help Show this message and exit.
terradev lineage export▾
Explainer coming soon — add a paragraph under this heading in CLI.md to replace this placeholder.
Usage: terradev lineage export [OPTIONS]
Export lineage data for compliance reports
Options:
--format [json|csv] Export format
--model TEXT Filter by model name
--env [dev|staging|prod] Filter by environment
-o, --output TEXT Output file (default: stdout)
--help Show this message and exit.
terradev lineage graph▾
Explainer coming soon — add a paragraph under this heading in CLI.md to replace this placeholder.
Usage: terradev lineage graph [OPTIONS] ARTIFACT_ID
Show lineage graph for artifact
Options:
--direction [up|down|both]
--help Show this message and exit.
terradev lineage production▾
Explainer coming soon — add a paragraph under this heading in CLI.md to replace this placeholder.
Usage: terradev lineage production [OPTIONS]
Show artifacts in production environment
Options:
--type TEXT Filter by artifact type
--help Show this message and exit.
terradev lineage register▾
Explainer coming soon — add a paragraph under this heading in CLI.md to replace this placeholder.
Usage: terradev lineage register [OPTIONS]
{dataset|model|checkpoint|metrics|config}
NAME URI
Register a new artifact for lineage tracking
Options:
--env [dev|staging|prod]
--hash TEXT Artifact hash
--size INTEGER Size in bytes
--user TEXT User registering artifact
--parent TEXT Parent artifact ID
--help Show this message and exit.
terradev lineage show▾
Explainer coming soon — add a paragraph under this heading in CLI.md to replace this placeholder.
Usage: terradev lineage show [OPTIONS] MODEL_IDENTIFIER
Show complete provenance of a model (auto-generated)
Options:
--env [dev|staging|prod] Filter by environment
--help Show this message and exit.
terradev lineage trace▾
Explainer coming soon — add a paragraph under this heading in CLI.md to replace this placeholder.
Usage: terradev lineage trace [OPTIONS]
Trace complete lineage from checkpoint or execution
Options:
--checkpoint TEXT Checkpoint ID to trace backwards from
--execution TEXT Execution ID to trace
--help Show this message and exit.
terradev local▾
Explainer coming soon — add a paragraph under this heading in CLI.md to replace this placeholder.
Usage: terradev local [OPTIONS] COMMAND [ARGS]...
Local GPU discovery and hybrid compute pool management.
Discover GPUs on this machine or remote hosts via SSH, register them into
your compute pool alongside cloud providers, and route workloads to the
cheapest available compute including $0/hr local hardware.
Options:
--help Show this message and exit.
Commands:
pool View or manage your hybrid compute pool (local + cloud...
register Register a local or remote GPU host into your compute pool.
scan Scan local machine or remote host for GPUs.
terradev local pool▾
Explainer coming soon — add a paragraph under this heading in CLI.md to replace this placeholder.
Usage: terradev local pool [OPTIONS]
View or manage your hybrid compute pool (local + cloud instances).
Shows all registered local/remote GPU hosts alongside active cloud
instances.
Example:
terradev local pool
terradev local pool --format json
terradev local pool --remove workstation-4090
Options:
--format [table|json] Output format
--remove TEXT Remove a pool entry by name
--help Show this message and exit.
terradev local register▾
Explainer coming soon — add a paragraph under this heading in CLI.md to replace this placeholder.
Usage: terradev local register [OPTIONS]
Register a local or remote GPU host into your compute pool.
Example:
terradev local register --name workstation-4090
terradev local register --name lab-node-01 --host 10.0.0.5 --user ubuntu
Options:
--name TEXT Name for this pool entry [required]
--host TEXT Remote host (omit for localhost)
--user TEXT SSH username for remote host
--key TEXT SSH private key path
--help Show this message and exit.
terradev local scan▾
Explainer coming soon — add a paragraph under this heading in CLI.md to replace this placeholder.
Usage: terradev local scan [OPTIONS]
Scan local machine or remote host for GPUs.
Uses Rust NVML bindings (5-10x faster than nvidia-smi) with automatic
fallback to nvidia-smi parsing if the Rust extension is unavailable.
Examples:
terradev local scan
terradev local scan --detailed
terradev local scan --host 192.168.1.50 --user ubuntu --key
~/.ssh/id_rsa
terradev local scan --register --name workstation-4090
Options:
--host TEXT Remote host IP/hostname to scan via SSH
--user TEXT SSH username for remote scan
--key TEXT Path to SSH private key for remote scan
--detailed Show full topology, PCIe, NUMA, clock details
--register Auto-register discovered GPUs into pool
--name TEXT Name for registered pool entry (auto-generated if omitted)
--help Show this message and exit.
terradev lora▾
Explainer coming soon — add a paragraph under this heading in CLI.md to replace this placeholder.
Usage: terradev lora [OPTIONS] COMMAND [ARGS]...
Production-grade LoRA adapter management with registry and cross-replica
consistency.
Manage adapter versions, track replica distribution, and ensure consistency
across deployments.
Options:
--help Show this message and exit.
Commands:
activate Activate a specific version across all replicas.
add Hot-load a LoRA adapter onto a running vLLM server.
cost-report Generate cost attribution report for LoRA adapters.
drift-check Check for performance drift in an adapter.
list List loaded LoRA adapters.
lorax LoRAX (LoRA eXchange) multi-LoRA inference server from...
peft HuggingFace PEFT adapter import and management.
register Register a LoRA adapter in the central registry with...
remove Hot-unload a LoRA adapter.
rollback Rollback adapter to previous stable version.
sync Synchronize adapter state across all replicas in a...
versions List all versions of an adapter.
terradev lora activate▾
Explainer coming soon — add a paragraph under this heading in CLI.md to replace this placeholder.
Usage: terradev lora activate [OPTIONS]
Activate a specific version across all replicas.
Examples: terradev lora activate -n customer-a -v abc123...
Options:
-n, --name TEXT Adapter name [required]
-v, --version TEXT Version ID to activate [required]
--help Show this message and exit.
terradev lora add▾
Explainer coming soon — add a paragraph under this heading in CLI.md to replace this placeholder.
Usage: terradev lora add [OPTIONS]
Hot-load a LoRA adapter onto a running vLLM server.
Examples: terradev lora add -e http://10.0.0.1:8000 -n customer-a --path
/adapters/customer-a terradev lora add -e http://10.0.0.1:8000 -n
customer-a --path /adapters/customer-a --register --base-model meta-
llama/Llama-2-7b-hf
Options:
-e, --endpoint TEXT vLLM endpoint [required]
-n, --name TEXT Adapter name (becomes the model name in API requests)
[required]
--path TEXT Path to adapter weights [required]
--api-key TEXT vLLM API key
--register Also register in central registry
--base-model TEXT Base model (required with --register)
--rank INTEGER LoRA rank (default: 64)
--help Show this message and exit.
terradev lora cost-report▾
Explainer coming soon — add a paragraph under this heading in CLI.md to replace this placeholder.
Usage: terradev lora cost-report [OPTIONS]
Generate cost attribution report for LoRA adapters.
Examples: terradev lora cost-report -d 7 terradev lora cost-report
-a customer-a terradev lora cost-report -t tenant-123
Options:
-d, --days INTEGER Number of days to report (default: 30)
-a, --adapter TEXT Specific adapter to report on
-t, --tenant TEXT Specific tenant to report on
--help Show this message and exit.
terradev lora drift-check▾
Explainer coming soon — add a paragraph under this heading in CLI.md to replace this placeholder.
Usage: terradev lora drift-check [OPTIONS]
Check for performance drift in an adapter.
Examples: terradev lora drift-check -n customer-a terradev lora
drift-check -n customer-a -t 0.15
Options:
-n, --name TEXT Adapter name to check [required]
-v, --version TEXT Specific version to check (default: active)
-t, --threshold FLOAT Drift threshold (default: 0.1)
--source TEXT Data source for drift detection
--help Show this message and exit.
terradev lora list▾
Explainer coming soon — add a paragraph under this heading in CLI.md to replace this placeholder.
Usage: terradev lora list [OPTIONS]
List loaded LoRA adapters.
Examples: terradev lora list -e http://10.0.0.1:8000 terradev lora
list -e http://10.0.0.1:8000 --registry
Options:
-e, --endpoint TEXT vLLM endpoint (e.g. http://10.0.0.1:8000) [required]
--api-key TEXT vLLM API key
--registry Show registry state instead of live endpoint
--help Show this message and exit.
terradev lora lorax▾
Explainer coming soon — add a paragraph under this heading in CLI.md to replace this placeholder.
Usage: terradev lora lorax [OPTIONS] COMMAND [ARGS]...
LoRAX (LoRA eXchange) multi-LoRA inference server from Predibase.
Deploy and manage LoRAX servers for serving thousands of fine-tuned models
on a single GPU with dynamic adapter loading.
Options:
--help Show this message and exit.
Commands:
deploy Deploy a LoRAX server.
generate Generate text using LoRAX server.
list-adapters List loaded adapters on LoRAX server.
load-adapter Load a LoRA adapter onto LoRAX server.
sync-registry Sync Terradev LoRA registry with LoRAX server state.
test Test LoRAX server connectivity.
unload-adapter Unload a LoRA adapter from LoRAX server.
terradev lora lorax deploy▾
Explainer coming soon — add a paragraph under this heading in CLI.md to replace this placeholder.
Usage: terradev lora lorax deploy [OPTIONS]
Deploy a LoRAX server.
Examples: terradev lora lorax deploy -m
mistralai/Mistral-7B-Instruct-v0.1 --docker terradev lora lorax deploy
-m meta-llama/Llama-2-7b-hf --k8s --namespace lorax
Options:
-m, --model-id TEXT Base model ID (e.g.,
mistralai/Mistral-7B-Instruct-v0.1)
[required]
--host TEXT LoRAX server host
-p, --port INTEGER LoRAX server port
--quantization [none|bitsandbytes|gptq|awq]
Quantization method
--gpu-memory-fraction FLOAT GPU memory fraction to use
--max-loras INTEGER Maximum number of adapters to load
--docker Deploy using Docker
--k8s Deploy using Kubernetes
--namespace TEXT Kubernetes namespace (for --k8s)
--help Show this message and exit.
terradev lora lorax generate▾
Explainer coming soon — add a paragraph under this heading in CLI.md to replace this placeholder.
Usage: terradev lora lorax generate [OPTIONS]
Generate text using LoRAX server.
Examples: terradev lora lorax generate -p "Hello, world!" terradev
lora lorax generate -p "What is 2+2?" -a my-adapter
Options:
-p, --prompt TEXT Input prompt [required]
-a, --adapter-id TEXT Adapter ID to use
--max-tokens INTEGER Max tokens to generate
--temperature FLOAT Sampling temperature
--host TEXT LoRAX server host
--port INTEGER LoRAX server port
--help Show this message and exit.
terradev lora lorax list-adapters▾
Explainer coming soon — add a paragraph under this heading in CLI.md to replace this placeholder.
Usage: terradev lora lorax list-adapters [OPTIONS]
List loaded adapters on LoRAX server.
Examples: terradev lora lorax list-adapters
Options:
--host TEXT LoRAX server host
-p, --port INTEGER LoRAX server port
--help Show this message and exit.
terradev lora lorax load-adapter▾
Explainer coming soon — add a paragraph under this heading in CLI.md to replace this placeholder.
Usage: terradev lora lorax load-adapter [OPTIONS]
Load a LoRA adapter onto LoRAX server.
Examples: terradev lora lorax load-adapter -a vineetsharma/qlora-
adapter-Mistral-7B-Instruct-v0.1-gsm8k terradev lora lorax load-adapter
-a /path/to/local/adapter --adapter-name my-adapter
Options:
-a, --adapter-id TEXT Adapter ID (HuggingFace repo or local path)
[required]
--adapter-name TEXT Custom name for the adapter
--host TEXT LoRAX server host
-p, --port INTEGER LoRAX server port
--help Show this message and exit.
terradev lora lorax sync-registry▾
Explainer coming soon — add a paragraph under this heading in CLI.md to replace this placeholder.
Usage: terradev lora lorax sync-registry [OPTIONS]
Sync Terradev LoRA registry with LoRAX server state.
Examples: terradev lora lorax sync-registry terradev lora lorax
sync-registry -a customer-a
Options:
--host TEXT LoRAX server host
-p, --port INTEGER LoRAX server port
-a, --adapter TEXT Specific adapter to sync
--help Show this message and exit.
terradev lora lorax test▾
Explainer coming soon — add a paragraph under this heading in CLI.md to replace this placeholder.
Usage: terradev lora lorax test [OPTIONS]
Test LoRAX server connectivity.
Examples: terradev lora lorax test terradev lora lorax test --host
10.0.0.1 --port 8080
Options:
--host TEXT LoRAX server host
-p, --port INTEGER LoRAX server port
--help Show this message and exit.
terradev lora lorax unload-adapter▾
Explainer coming soon — add a paragraph under this heading in CLI.md to replace this placeholder.
Usage: terradev lora lorax unload-adapter [OPTIONS]
Unload a LoRA adapter from LoRAX server.
Examples: terradev lora lorax unload-adapter -a vineetsharma/qlora-
adapter-Mistral-7B-Instruct-v0.1-gsm8k
Options:
-a, --adapter-id TEXT Adapter ID to unload [required]
--host TEXT LoRAX server host
-p, --port INTEGER LoRAX server port
--help Show this message and exit.
terradev lora peft▾
Explainer coming soon — add a paragraph under this heading in CLI.md to replace this placeholder.
Usage: terradev lora peft [OPTIONS] COMMAND [ARGS]...
HuggingFace PEFT adapter import and management.
Download, validate, and prepare LoRA adapters from HuggingFace for use with
vLLM, LoRAX, or other inference servers.
Options:
--help Show this message and exit.
Commands:
delete Delete a locally imported adapter.
import Import a LoRA adapter from HuggingFace.
list List all locally imported PEFT adapters.
validate Validate a PEFT adapter structure.
terradev lora peft delete▾
Explainer coming soon — add a paragraph under this heading in CLI.md to replace this placeholder.
Usage: terradev lora peft delete [OPTIONS]
Delete a locally imported adapter.
Examples: terradev lora peft delete -a username/adapter-name
Options:
-a, --adapter-id TEXT Adapter ID to delete [required]
--help Show this message and exit.
terradev lora peft import▾
Explainer coming soon — add a paragraph under this heading in CLI.md to replace this placeholder.
Usage: terradev lora peft import [OPTIONS]
Import a LoRA adapter from HuggingFace.
Examples: terradev lora peft import -a vineetsharma/qlora-adapter-
Mistral-7B-Instruct-v0.1-gsm8k terradev lora peft import -a
username/adapter --local-name my-adapter --register --base-model
mistralai/Mistral-7B-Instruct-v0.1
Options:
-a, --adapter-id TEXT HuggingFace adapter ID (e.g., username/adapter-name)
[required]
--local-name TEXT Local name for the adapter
--token TEXT HuggingFace auth token (for private repos)
--register Register imported adapter in Terradev registry
-b, --base-model TEXT Base model (required with --register)
--rank INTEGER LoRA rank (auto-detected if not specified)
--help Show this message and exit.
terradev lora peft list▾
Explainer coming soon — add a paragraph under this heading in CLI.md to replace this placeholder.
Usage: terradev lora peft list [OPTIONS]
List all locally imported PEFT adapters.
Examples: terradev lora peft list
Options:
--help Show this message and exit.
terradev lora peft validate▾
Explainer coming soon — add a paragraph under this heading in CLI.md to replace this placeholder.
Usage: terradev lora peft validate [OPTIONS]
Validate a PEFT adapter structure.
Examples: terradev lora peft validate -p
~/.terradev/peft_adapters/username--adapter-name
Options:
-p, --path TEXT Path to adapter directory [required]
--help Show this message and exit.
terradev lora register▾
Explainer coming soon — add a paragraph under this heading in CLI.md to replace this placeholder.
Usage: terradev lora register [OPTIONS]
Register a LoRA adapter in the central registry with version tracking.
Examples: terradev lora register -n customer-a --path
/adapters/customer-a -b meta-llama/Llama-2-7b-hf terradev lora register
-n customer-b --path /adapters/customer-b -b meta-llama/Llama-2-7b-hf
--tenant t-123
Options:
-n, --name TEXT Adapter name [required]
--path TEXT Path to adapter weights [required]
-b, --base-model TEXT Base model name (e.g., meta-llama/Llama-2-7b-hf)
[required]
--rank INTEGER LoRA rank (default: 64)
--tenant TEXT Associate with tenant ID
--metadata TEXT JSON metadata string
--help Show this message and exit.
terradev lora remove▾
Explainer coming soon — add a paragraph under this heading in CLI.md to replace this placeholder.
Usage: terradev lora remove [OPTIONS]
Hot-unload a LoRA adapter.
Examples: terradev lora remove -e http://10.0.0.1:8000 -n customer-a
Options:
-e, --endpoint TEXT vLLM endpoint [required]
-n, --name TEXT Adapter name to unload [required]
--api-key TEXT vLLM API key
--help Show this message and exit.
terradev lora rollback▾
Explainer coming soon — add a paragraph under this heading in CLI.md to replace this placeholder.
Usage: terradev lora rollback [OPTIONS]
Rollback adapter to previous stable version.
Examples: terradev lora rollback -n customer-a terradev lora
rollback -n customer-a -v abc123... --replicas 10.0.0.1:8000,10.0.0.2:8000
Options:
-n, --name TEXT Adapter name to rollback [required]
-v, --to-version TEXT Target version ID (default: previous stable)
--replicas TEXT Comma-separated list of replica endpoints (host:port)
--help Show this message and exit.
terradev lora sync▾
Explainer coming soon — add a paragraph under this heading in CLI.md to replace this placeholder.
Usage: terradev lora sync [OPTIONS]
Synchronize adapter state across all replicas in a deployment.
Examples: terradev lora sync -d prod -n customer-a --replicas
10.0.0.1:8000,10.0.0.2:8000
Options:
-d, --deployment TEXT Deployment name [required]
-n, --name TEXT Adapter name [required]
--replicas TEXT Comma-separated list of replica endpoints (host:port)
--help Show this message and exit.
terradev lora versions▾
Explainer coming soon — add a paragraph under this heading in CLI.md to replace this placeholder.
Usage: terradev lora versions [OPTIONS]
List all versions of an adapter.
Examples: terradev lora versions -n customer-a
Options:
-n, --name TEXT Adapter name [required]
--help Show this message and exit.
terradev manage▾
Explainer coming soon — add a paragraph under this heading in CLI.md to replace this placeholder.
Usage: terradev manage [OPTIONS]
Manage provisioned GPU instances via provider APIs.
Control the lifecycle of your GPU instances by checking status, stopping,
starting, or terminating them. Actions are sent directly to the cloud
provider.
Examples: terradev manage -i <instance-id> -a status # Check instance
status terradev manage -i <instance-id> -a stop # Stop instance
(keeps allocation) terradev manage -i <instance-id> -a start # Start
stopped instance terradev manage -i <instance-id> -a terminate #
Terminate and release
Actions: - status: Query provider for current instance status - stop:
Stop the instance (keeps allocation, you pay for storage) - start: Start a
stopped instance - terminate: Permanently terminate and release resources
Instance IDs: Get instance IDs from: terradev status Use the full ID
shown in the status output
Cost Implications: - stop: You may still pay for storage/allocation
depending on provider - terminate: No further charges after termination
Options:
-i, --instance-id TEXT Instance ID (from terradev status)
[required]
-a, --action [status|stop|start|terminate]
Action: status (default), stop, start,
terminate
--help Show this message and exit.
terradev manifests▾
Explainer coming soon — add a paragraph under this heading in CLI.md to replace this placeholder.
Usage: terradev manifests [OPTIONS]
List cached manifests, imported pipelines, and recordings
Options:
--job TEXT Show versions for specific job
--cache-dir TEXT Manifest cache directory
--show-imported Show imported YAML pipelines
--show-recordings Show live recordings
--help Show this message and exit.
terradev mcp▾
Explainer coming soon — add a paragraph under this heading in CLI.md to replace this placeholder.
Usage: terradev mcp [OPTIONS] {serve|install|list-tools}
Run Terradev as an MCP server for agent integration.
Makes Terradev callable from AI agents (Claude Desktop, Cursor, Windsurf,
Continue, Cline).
Actions: serve: Start MCP server (default: stdio transport) install:
Install MCP config for a specific client list-tools: List all available
MCP tools
Options:
--client [claude-desktop|cursor|windsurf|continue|cline]
Client to install MCP config for
--transport [stdio|sse|http] MCP transport protocol
--help Show this message and exit.
terradev migrate▾
Explainer coming soon — add a paragraph under this heading in CLI.md to replace this placeholder.
Usage: terradev migrate [OPTIONS] COMMAND [ARGS]...
Cross-provider workload migration with dry-run analysis
Options:
--help Show this message and exit.
Commands:
list-workloads List available workloads for migration
migration Migrate workload between providers with detailed cost...
terradev migrate list-workloads▾
Explainer coming soon — add a paragraph under this heading in CLI.md to replace this placeholder.
Usage: terradev migrate list-workloads [OPTIONS]
List available workloads for migration
Options:
--provider TEXT Filter by provider
--format [table|json]
--help Show this message and exit.
terradev migrate migration▾
Explainer coming soon — add a paragraph under this heading in CLI.md to replace this placeholder.
Usage: terradev migrate migration [OPTIONS]
Migrate workload between providers with detailed cost analysis
Options:
--from TEXT Source provider [required]
--to TEXT Target provider [required]
--instance-id TEXT Source instance ID
--workload TEXT Workload ID from JobStateManager
--dry-run Show migration plan without executing
--help Show this message and exit.
terradev ml▾
Explainer coming soon — add a paragraph under this heading in CLI.md to replace this placeholder.
Usage: terradev ml [OPTIONS] COMMAND [ARGS]...
ML Platform Integration Commands
Options:
--help Show this message and exit.
Commands:
databricks Databricks MLOps jobs, clusters, model serving, MLflow.
dvc DVC (Data Version Control) management.
guardrails NeMo Guardrails LLM output safety, jailbreak detection,...
kserve KServe model deployment and management.
langchain LangChain integration with workflows, LangGraph, and...
langfuse Langfuse LLM observability traces, scores, datasets,...
langgraph LangGraph workflow orchestration with monitoring.
langsmith LangSmith experiment tracking and monitoring.
mlflow-legacy MLflow experiment tracking and model registry.
phoenix Arize Phoenix LLM trace observability traces, spans, OTEL.
qdrant Qdrant vector database collections, search, RAG...
ray Enhanced Ray distributed computing with monitoring and...
sglang SGLang optimization and management with...
vllm vLLM optimization and management commands.
wandb Weights & Biases experiment tracking with dashboards,...
terradev ml databricks▾
Explainer coming soon — add a paragraph under this heading in CLI.md to replace this placeholder.
Usage: terradev ml databricks [OPTIONS] COMMAND [ARGS]...
Databricks MLOps jobs, clusters, model serving, MLflow.
Options:
--help Show this message and exit.
Commands:
clusters List Databricks clusters.
configure Configure Databricks credentials.
deploy-model Deploy a model to a serving endpoint.
jobs List Databricks jobs.
mlflow Databricks-hosted MLflow operations.
query Query a model serving endpoint.
run Trigger a Databricks job run.
run-status Get status of a Databricks run.
serving-endpoints List model serving endpoints.
test Test Databricks connectivity.
terradev ml databricks clusters▾
Explainer coming soon — add a paragraph under this heading in CLI.md to replace this placeholder.
Usage: terradev ml databricks clusters [OPTIONS]
List Databricks clusters.
Options:
-f, --format [json|text]
--help Show this message and exit.
terradev ml databricks configure▾
Explainer coming soon — add a paragraph under this heading in CLI.md to replace this placeholder.
Usage: terradev ml databricks configure [OPTIONS]
Configure Databricks credentials.
Options:
--host TEXT
--token TEXT
--help Show this message and exit.
terradev ml databricks deploy-model▾
Explainer coming soon — add a paragraph under this heading in CLI.md to replace this placeholder.
Usage: terradev ml databricks deploy-model [OPTIONS]
Deploy a model to a serving endpoint.
Options:
--endpoint-name TEXT Serving endpoint name [required]
--model-name TEXT Registered model name [required]
--model-version TEXT Model version
--workload-size [Small|Medium|Large]
--scale-to-zero / --no-scale-to-zero
-f, --format [json|text]
--help Show this message and exit.
terradev ml databricks jobs▾
Explainer coming soon — add a paragraph under this heading in CLI.md to replace this placeholder.
Usage: terradev ml databricks jobs [OPTIONS]
List Databricks jobs.
Options:
-n, --limit INTEGER
-f, --format [json|text]
--help Show this message and exit.
terradev ml databricks mlflow▾
Explainer coming soon — add a paragraph under this heading in CLI.md to replace this placeholder.
Usage: terradev ml databricks mlflow [OPTIONS] COMMAND [ARGS]...
Databricks-hosted MLflow operations.
Options:
--help Show this message and exit.
Commands:
experiments List MLflow experiments.
models List registered models in Databricks Model Registry.
terradev ml databricks mlflow experiments▾
Explainer coming soon — add a paragraph under this heading in CLI.md to replace this placeholder.
Usage: terradev ml databricks mlflow experiments [OPTIONS]
List MLflow experiments.
Options:
-n, --limit INTEGER
-f, --format [json|text]
--help Show this message and exit.
terradev ml databricks mlflow models▾
Explainer coming soon — add a paragraph under this heading in CLI.md to replace this placeholder.
Usage: terradev ml databricks mlflow models [OPTIONS]
List registered models in Databricks Model Registry.
Options:
-n, --limit INTEGER
-f, --format [json|text]
--help Show this message and exit.
terradev ml databricks query▾
Explainer coming soon — add a paragraph under this heading in CLI.md to replace this placeholder.
Usage: terradev ml databricks query [OPTIONS]
Query a model serving endpoint.
Options:
--endpoint TEXT Serving endpoint name [required]
--prompt TEXT Prompt text [required]
-f, --format [json|text]
--help Show this message and exit.
terradev ml databricks run▾
Explainer coming soon — add a paragraph under this heading in CLI.md to replace this placeholder.
Usage: terradev ml databricks run [OPTIONS] JOB_ID
Trigger a Databricks job run.
Options:
-f, --format [json|text]
--help Show this message and exit.
terradev ml databricks run-status▾
Explainer coming soon — add a paragraph under this heading in CLI.md to replace this placeholder.
Usage: terradev ml databricks run-status [OPTIONS] RUN_ID
Get status of a Databricks run.
Options:
-f, --format [json|text]
--help Show this message and exit.
terradev ml databricks serving-endpoints▾
Explainer coming soon — add a paragraph under this heading in CLI.md to replace this placeholder.
Usage: terradev ml databricks serving-endpoints [OPTIONS]
List model serving endpoints.
Options:
-f, --format [json|text]
--help Show this message and exit.
terradev ml databricks test▾
Explainer coming soon — add a paragraph under this heading in CLI.md to replace this placeholder.
Usage: terradev ml databricks test [OPTIONS]
Test Databricks connectivity.
Options:
--help Show this message and exit.
terradev ml dvc▾
Explainer coming soon — add a paragraph under this heading in CLI.md to replace this placeholder.
Usage: terradev ml dvc [OPTIONS] COMMAND [ARGS]...
DVC (Data Version Control) management.
Options:
--help Show this message and exit.
Commands:
add-data Add data to tracking.
add-remote Add remote storage (name:url).
init Initialize DVC repository.
pull Pull data from remote.
push Push data to remote.
status Show repository status.
test Test connection to DVC service.
terradev ml dvc add-data▾
Explainer coming soon — add a paragraph under this heading in CLI.md to replace this placeholder.
Usage: terradev ml dvc add-data [OPTIONS] DATA_PATH
Add data to tracking.
Options:
--help Show this message and exit.
terradev ml dvc add-remote▾
Explainer coming soon — add a paragraph under this heading in CLI.md to replace this placeholder.
Usage: terradev ml dvc add-remote [OPTIONS] REMOTE_SPEC
Add remote storage (name:url).
Options:
--help Show this message and exit.
terradev ml dvc init▾
Explainer coming soon — add a paragraph under this heading in CLI.md to replace this placeholder.
Usage: terradev ml dvc init [OPTIONS]
Initialize DVC repository.
Options:
--help Show this message and exit.
terradev ml dvc pull▾
Explainer coming soon — add a paragraph under this heading in CLI.md to replace this placeholder.
Usage: terradev ml dvc pull [OPTIONS]
Pull data from remote.
Options:
--help Show this message and exit.
terradev ml dvc push▾
Explainer coming soon — add a paragraph under this heading in CLI.md to replace this placeholder.
Usage: terradev ml dvc push [OPTIONS]
Push data to remote.
Options:
--help Show this message and exit.
terradev ml dvc status▾
Explainer coming soon — add a paragraph under this heading in CLI.md to replace this placeholder.
Usage: terradev ml dvc status [OPTIONS]
Show repository status.
Options:
--help Show this message and exit.
terradev ml dvc test▾
Explainer coming soon — add a paragraph under this heading in CLI.md to replace this placeholder.
Usage: terradev ml dvc test [OPTIONS]
Test connection to DVC service.
Options:
--help Show this message and exit.
terradev ml guardrails▾
Explainer coming soon — add a paragraph under this heading in CLI.md to replace this placeholder.
Usage: terradev ml guardrails [OPTIONS] COMMAND [ARGS]...
NeMo Guardrails LLM output safety, jailbreak detection, PII masking.
Options:
--help Show this message and exit.
Commands:
chat Send a message through guardrails and show the result.
generate-config Generate default Colang 2.x guardrails configuration.
k8s Print K8s deployment manifest for guardrails server.
test Test connection to guardrails server.
terradev ml guardrails chat▾
Explainer coming soon — add a paragraph under this heading in CLI.md to replace this placeholder.
Usage: terradev ml guardrails chat [OPTIONS]
Send a message through guardrails and show the result.
Options:
-m, --message TEXT Message to send through guardrails [required]
-c, --config-id TEXT Guardrails config_id
--help Show this message and exit.
terradev ml guardrails generate-config▾
Explainer coming soon — add a paragraph under this heading in CLI.md to replace this placeholder.
Usage: terradev ml guardrails generate-config [OPTIONS]
Generate default Colang 2.x guardrails configuration.
Options:
-c, --config-id TEXT Config ID name
-o, --output-dir TEXT Output directory
--help Show this message and exit.
terradev ml guardrails k8s▾
Explainer coming soon — add a paragraph under this heading in CLI.md to replace this placeholder.
Usage: terradev ml guardrails k8s [OPTIONS]
Print K8s deployment manifest for guardrails server.
Options:
-n, --namespace TEXT K8s namespace
--help Show this message and exit.
terradev ml guardrails test▾
Explainer coming soon — add a paragraph under this heading in CLI.md to replace this placeholder.
Usage: terradev ml guardrails test [OPTIONS]
Test connection to guardrails server.
Options:
--help Show this message and exit.
terradev ml kserve▾
Explainer coming soon — add a paragraph under this heading in CLI.md to replace this placeholder.
Usage: terradev ml kserve [OPTIONS] COMMAND [ARGS]...
KServe model deployment and management.
Options:
--help Show this message and exit.
Commands:
test Test connection to KServe service.
terradev ml kserve test▾
Explainer coming soon — add a paragraph under this heading in CLI.md to replace this placeholder.
Usage: terradev ml kserve test [OPTIONS]
Test connection to KServe service.
Options:
--help Show this message and exit.
terradev ml langchain▾
Explainer coming soon — add a paragraph under this heading in CLI.md to replace this placeholder.
Usage: terradev ml langchain [OPTIONS] COMMAND [ARGS]...
LangChain integration with workflows, LangGraph, and SGLang.
Options:
--help Show this message and exit.
Commands:
create-langgraph Create a LangGraph workflow.
create-pipeline Create an SGLang pipeline.
create-trace Create a trace in LangSmith.
create-workflow Create a LangChain workflow.
list-projects List LangSmith projects.
list-runs List LangSmith runs.
test Test connection to LangChain service.
terradev ml langchain create-langgraph▾
Explainer coming soon — add a paragraph under this heading in CLI.md to replace this placeholder.
Usage: terradev ml langchain create-langgraph [OPTIONS] GRAPH_NAME
Create a LangGraph workflow.
Options:
--help Show this message and exit.
terradev ml langchain create-pipeline▾
Explainer coming soon — add a paragraph under this heading in CLI.md to replace this placeholder.
Usage: terradev ml langchain create-pipeline [OPTIONS] PIPELINE_NAME
Create an SGLang pipeline.
Options:
--help Show this message and exit.
terradev ml langchain create-trace▾
Explainer coming soon — add a paragraph under this heading in CLI.md to replace this placeholder.
Usage: terradev ml langchain create-trace [OPTIONS]
Create a trace in LangSmith.
Options:
-r, --run-id TEXT Run ID for trace [required]
-d, --data TEXT Trace data (JSON) [required]
--help Show this message and exit.
terradev ml langchain create-workflow▾
Explainer coming soon — add a paragraph under this heading in CLI.md to replace this placeholder.
Usage: terradev ml langchain create-workflow [OPTIONS] WORKFLOW_NAME
Create a LangChain workflow.
Options:
--help Show this message and exit.
terradev ml langchain list-projects▾
Explainer coming soon — add a paragraph under this heading in CLI.md to replace this placeholder.
Usage: terradev ml langchain list-projects [OPTIONS]
List LangSmith projects.
Options:
--help Show this message and exit.
terradev ml langchain list-runs▾
Explainer coming soon — add a paragraph under this heading in CLI.md to replace this placeholder.
Usage: terradev ml langchain list-runs [OPTIONS]
List LangSmith runs.
Options:
-p, --project TEXT LangSmith project name
--help Show this message and exit.
terradev ml langchain test▾
Explainer coming soon — add a paragraph under this heading in CLI.md to replace this placeholder.
Usage: terradev ml langchain test [OPTIONS]
Test connection to LangChain service.
Options:
--help Show this message and exit.
terradev ml langfuse▾
Explainer coming soon — add a paragraph under this heading in CLI.md to replace this placeholder.
Usage: terradev ml langfuse [OPTIONS] COMMAND [ARGS]...
Langfuse LLM observability traces, scores, datasets, prompts.
Options:
--help Show this message and exit.
Commands:
configure Configure Langfuse credentials.
datasets List datasets.
export-training-data Export traces as instruction/response pairs for...
k8s Print K8s deployment manifest for Langfuse.
otel-env Print OTEL env vars for instrumenting LLM apps.
quality Get quality metrics for drift detection.
score Create a score for a trace.
scores List scores.
test Test Langfuse connectivity.
trace Get a single trace with observations.
traces List recent traces.
terradev ml langfuse configure▾
Explainer coming soon — add a paragraph under this heading in CLI.md to replace this placeholder.
Usage: terradev ml langfuse configure [OPTIONS]
Configure Langfuse credentials.
Options:
--public-key TEXT
--secret-key TEXT
--host TEXT Langfuse server URL
--help Show this message and exit.
terradev ml langfuse datasets▾
Explainer coming soon — add a paragraph under this heading in CLI.md to replace this placeholder.
Usage: terradev ml langfuse datasets [OPTIONS]
List datasets.
Options:
-n, --limit INTEGER
-f, --format [json|text]
--help Show this message and exit.
terradev ml langfuse export-training-data▾
Explainer coming soon — add a paragraph under this heading in CLI.md to replace this placeholder.
Usage: terradev ml langfuse export-training-data [OPTIONS]
Export traces as instruction/response pairs for LoRA fine-tuning.
Options:
-n, --limit INTEGER Max pairs to export
--name TEXT Filter traces by name
--min-score FLOAT Min quality score (0.0-1.0)
--score-name TEXT Score name to filter on
-o, --output TEXT Output file path (default: stdout)
--help Show this message and exit.
terradev ml langfuse k8s▾
Explainer coming soon — add a paragraph under this heading in CLI.md to replace this placeholder.
Usage: terradev ml langfuse k8s [OPTIONS]
Print K8s deployment manifest for Langfuse.
Options:
-n, --namespace TEXT K8s namespace
--help Show this message and exit.
terradev ml langfuse otel-env▾
Explainer coming soon — add a paragraph under this heading in CLI.md to replace this placeholder.
Usage: terradev ml langfuse otel-env [OPTIONS]
Print OTEL env vars for instrumenting LLM apps.
Options:
-p, --project TEXT Project name
--help Show this message and exit.
terradev ml langfuse quality▾
Explainer coming soon — add a paragraph under this heading in CLI.md to replace this placeholder.
Usage: terradev ml langfuse quality [OPTIONS]
Get quality metrics for drift detection.
Options:
--score-name TEXT Score name to aggregate
-n, --limit INTEGER
-f, --format [json|text]
--help Show this message and exit.
terradev ml langfuse score▾
Explainer coming soon — add a paragraph under this heading in CLI.md to replace this placeholder.
Usage: terradev ml langfuse score [OPTIONS]
Create a score for a trace.
Options:
--trace-id TEXT Trace to score [required]
--name TEXT Score name (e.g. accuracy, quality) [required]
--value FLOAT Score value (numeric) [required]
--observation-id TEXT Specific observation to score
--comment TEXT Optional comment
--help Show this message and exit.
terradev ml langfuse scores▾
Explainer coming soon — add a paragraph under this heading in CLI.md to replace this placeholder.
Usage: terradev ml langfuse scores [OPTIONS]
List scores.
Options:
--trace-id TEXT Filter by trace ID
--name TEXT Filter by score name
-n, --limit INTEGER
-f, --format [json|text]
--help Show this message and exit.
terradev ml langfuse test▾
Explainer coming soon — add a paragraph under this heading in CLI.md to replace this placeholder.
Usage: terradev ml langfuse test [OPTIONS]
Test Langfuse connectivity.
Options:
--help Show this message and exit.
terradev ml langfuse trace▾
Explainer coming soon — add a paragraph under this heading in CLI.md to replace this placeholder.
Usage: terradev ml langfuse trace [OPTIONS] TRACE_ID
Get a single trace with observations.
Options:
-f, --format [json|text]
--help Show this message and exit.
terradev ml langfuse traces▾
Explainer coming soon — add a paragraph under this heading in CLI.md to replace this placeholder.
Usage: terradev ml langfuse traces [OPTIONS]
List recent traces.
Options:
-n, --limit INTEGER
--name TEXT Filter by trace name
-f, --format [json|text]
--help Show this message and exit.
terradev ml langgraph▾
Explainer coming soon — add a paragraph under this heading in CLI.md to replace this placeholder.
Usage: terradev ml langgraph [OPTIONS] COMMAND [ARGS]...
LangGraph workflow orchestration with monitoring.
Options:
--help Show this message and exit.
Commands:
create-workflow Create a LangGraph workflow.
deploy Deploy a workflow.
status Get workflow status.
test Test connection to LangGraph service.
terradev ml langgraph create-workflow▾
Explainer coming soon — add a paragraph under this heading in CLI.md to replace this placeholder.
Usage: terradev ml langgraph create-workflow [OPTIONS] WORKFLOW_NAME
Create a LangGraph workflow.
Options:
-t, --type [orchestrator-worker|evaluator-optimizer]
Workflow type [required]
--help Show this message and exit.
terradev ml langgraph deploy▾
Explainer coming soon — add a paragraph under this heading in CLI.md to replace this placeholder.
Usage: terradev ml langgraph deploy [OPTIONS] WORKFLOW_NAME
Deploy a workflow.
Options:
--help Show this message and exit.
terradev ml langgraph status▾
Explainer coming soon — add a paragraph under this heading in CLI.md to replace this placeholder.
Usage: terradev ml langgraph status [OPTIONS] WORKFLOW_ID
Get workflow status.
Options:
--help Show this message and exit.
terradev ml langgraph test▾
Explainer coming soon — add a paragraph under this heading in CLI.md to replace this placeholder.
Usage: terradev ml langgraph test [OPTIONS]
Test connection to LangGraph service.
Options:
--help Show this message and exit.
terradev ml langsmith▾
Explainer coming soon — add a paragraph under this heading in CLI.md to replace this placeholder.
Usage: terradev ml langsmith [OPTIONS] COMMAND [ARGS]...
LangSmith experiment tracking and monitoring.
Options:
--help Show this message and exit.
Commands:
create-project Create a new LangSmith project.
export Export runs data.
list-projects List all LangSmith projects.
test Test connection to LangSmith service.
terradev ml langsmith create-project▾
Explainer coming soon — add a paragraph under this heading in CLI.md to replace this placeholder.
Usage: terradev ml langsmith create-project [OPTIONS] PROJECT_NAME
Create a new LangSmith project.
Options:
--help Show this message and exit.
terradev ml langsmith export▾
Explainer coming soon — add a paragraph under this heading in CLI.md to replace this placeholder.
Usage: terradev ml langsmith export [OPTIONS]
Export runs data.
Options:
-f, --format [json|csv] Export format
--help Show this message and exit.
terradev ml langsmith list-projects▾
Explainer coming soon — add a paragraph under this heading in CLI.md to replace this placeholder.
Usage: terradev ml langsmith list-projects [OPTIONS]
List all LangSmith projects.
Options:
--help Show this message and exit.
terradev ml langsmith test▾
Explainer coming soon — add a paragraph under this heading in CLI.md to replace this placeholder.
Usage: terradev ml langsmith test [OPTIONS]
Test connection to LangSmith service.
Options:
--help Show this message and exit.
terradev ml mlflow-legacy▾
Explainer coming soon — add a paragraph under this heading in CLI.md to replace this placeholder.
Usage: terradev ml mlflow-legacy [OPTIONS] COMMAND [ARGS]...
MLflow experiment tracking and model registry.
Options:
--help Show this message and exit.
Commands:
create-experiment Create a new MLflow experiment.
export Export experiment data.
list-experiments List all MLflow experiments.
list-runs List runs in experiment.
test Test connection to MLflow service.
terradev ml mlflow-legacy create-experiment▾
Explainer coming soon — add a paragraph under this heading in CLI.md to replace this placeholder.
Usage: terradev ml mlflow-legacy create-experiment [OPTIONS] EXPERIMENT_NAME
Create a new MLflow experiment.
Options:
--help Show this message and exit.
terradev ml mlflow-legacy export▾
Explainer coming soon — add a paragraph under this heading in CLI.md to replace this placeholder.
Usage: terradev ml mlflow-legacy export [OPTIONS] EXPERIMENT_ID
Export experiment data.
Options:
-f, --format [json|csv] Export format
--help Show this message and exit.
terradev ml mlflow-legacy list-experiments▾
Explainer coming soon — add a paragraph under this heading in CLI.md to replace this placeholder.
Usage: terradev ml mlflow-legacy list-experiments [OPTIONS]
List all MLflow experiments.
Options:
--help Show this message and exit.
terradev ml mlflow-legacy list-runs▾
Explainer coming soon — add a paragraph under this heading in CLI.md to replace this placeholder.
Usage: terradev ml mlflow-legacy list-runs [OPTIONS] EXPERIMENT_ID
List runs in experiment.
Options:
--help Show this message and exit.
terradev ml mlflow-legacy test▾
Explainer coming soon — add a paragraph under this heading in CLI.md to replace this placeholder.
Usage: terradev ml mlflow-legacy test [OPTIONS]
Test connection to MLflow service.
Options:
--help Show this message and exit.
terradev ml phoenix▾
Explainer coming soon — add a paragraph under this heading in CLI.md to replace this placeholder.
Usage: terradev ml phoenix [OPTIONS] COMMAND [ARGS]...
Arize Phoenix LLM trace observability traces, spans, OTEL.
Options:
--help Show this message and exit.
Commands:
k8s Print K8s deployment manifest for Phoenix server.
otel-env Print OTEL env vars to inject into serving pods.
projects List Phoenix projects.
snippet Print Python instrumentation snippet.
spans List recent spans for a project.
test Test connection to Phoenix server.
trace View full execution tree for a trace.
terradev ml phoenix k8s▾
Explainer coming soon — add a paragraph under this heading in CLI.md to replace this placeholder.
Usage: terradev ml phoenix k8s [OPTIONS]
Print K8s deployment manifest for Phoenix server.
Options:
-n, --namespace TEXT K8s namespace
--help Show this message and exit.
terradev ml phoenix otel-env▾
Explainer coming soon — add a paragraph under this heading in CLI.md to replace this placeholder.
Usage: terradev ml phoenix otel-env [OPTIONS]
Print OTEL env vars to inject into serving pods.
Options:
-p, --project TEXT Project name
--help Show this message and exit.
terradev ml phoenix projects▾
Explainer coming soon — add a paragraph under this heading in CLI.md to replace this placeholder.
Usage: terradev ml phoenix projects [OPTIONS]
List Phoenix projects.
Options:
-l, --limit INTEGER Max projects to return
--help Show this message and exit.
terradev ml phoenix snippet▾
Explainer coming soon — add a paragraph under this heading in CLI.md to replace this placeholder.
Usage: terradev ml phoenix snippet [OPTIONS]
Print Python instrumentation snippet.
Options:
-p, --project TEXT Project name
--help Show this message and exit.
terradev ml phoenix spans▾
Explainer coming soon — add a paragraph under this heading in CLI.md to replace this placeholder.
Usage: terradev ml phoenix spans [OPTIONS]
List recent spans for a project.
Options:
-p, --project TEXT Project ID or name
-f, --filter TEXT SpanQuery DSL filter, e.g. "span_kind == 'RETRIEVER'"
-l, --limit INTEGER Max spans
--help Show this message and exit.
terradev ml phoenix test▾
Explainer coming soon — add a paragraph under this heading in CLI.md to replace this placeholder.
Usage: terradev ml phoenix test [OPTIONS]
Test connection to Phoenix server.
Options:
--help Show this message and exit.
terradev ml phoenix trace▾
Explainer coming soon — add a paragraph under this heading in CLI.md to replace this placeholder.
Usage: terradev ml phoenix trace [OPTIONS]
View full execution tree for a trace.
Options:
-t, --trace-id TEXT Trace ID to inspect [required]
-p, --project TEXT Project ID or name
--help Show this message and exit.
terradev ml qdrant▾
Explainer coming soon — add a paragraph under this heading in CLI.md to replace this placeholder.
Usage: terradev ml qdrant [OPTIONS] COMMAND [ARGS]...
Qdrant vector database collections, search, RAG infrastructure.
Options:
--help Show this message and exit.
Commands:
collections List all collections.
count Count points in a collection.
create-collection Create a vector collection (auto-configured for...
info Get collection info and stats.
k8s Print K8s StatefulSet manifest for Qdrant.
test Test connection to Qdrant server.
terradev ml qdrant collections▾
Explainer coming soon — add a paragraph under this heading in CLI.md to replace this placeholder.
Usage: terradev ml qdrant collections [OPTIONS]
List all collections.
Options:
--help Show this message and exit.
terradev ml qdrant count▾
Explainer coming soon — add a paragraph under this heading in CLI.md to replace this placeholder.
Usage: terradev ml qdrant count [OPTIONS]
Count points in a collection.
Options:
-n, --name TEXT Collection name
--help Show this message and exit.
terradev ml qdrant create-collection▾
Explainer coming soon — add a paragraph under this heading in CLI.md to replace this placeholder.
Usage: terradev ml qdrant create-collection [OPTIONS]
Create a vector collection (auto-configured for embedding model).
Options:
-n, --name TEXT Collection name
-e, --embedding-model TEXT Embedding model (auto-sets vector size)
--help Show this message and exit.
terradev ml qdrant info▾
Explainer coming soon — add a paragraph under this heading in CLI.md to replace this placeholder.
Usage: terradev ml qdrant info [OPTIONS]
Get collection info and stats.
Options:
-n, --name TEXT Collection name
--help Show this message and exit.
terradev ml qdrant k8s▾
Explainer coming soon — add a paragraph under this heading in CLI.md to replace this placeholder.
Usage: terradev ml qdrant k8s [OPTIONS]
Print K8s StatefulSet manifest for Qdrant.
Options:
-n, --namespace TEXT K8s namespace
--help Show this message and exit.
terradev ml qdrant test▾
Explainer coming soon — add a paragraph under this heading in CLI.md to replace this placeholder.
Usage: terradev ml qdrant test [OPTIONS]
Test connection to Qdrant server.
Options:
--help Show this message and exit.
terradev ml ray▾
Explainer coming soon — add a paragraph under this heading in CLI.md to replace this placeholder.
Usage: terradev ml ray [OPTIONS] COMMAND [ARGS]...
Enhanced Ray distributed computing with monitoring and dashboards.
Options:
--help Show this message and exit.
Commands:
dashboard Get dashboard URL.
grafana Access Grafana dashboard.
install Show installation instructions.
install-monitoring Install monitoring stack with Ray dashboards.
list-nodes List cluster nodes.
metrics-summary Get comprehensive metrics summary.
prometheus Access Prometheus metrics.
start Start Ray cluster.
status Show cluster status.
stop Stop Ray cluster.
test Test connection to Ray service.
terradev ml ray dashboard▾
Explainer coming soon — add a paragraph under this heading in CLI.md to replace this placeholder.
Usage: terradev ml ray dashboard [OPTIONS]
Get dashboard URL.
Options:
--help Show this message and exit.
terradev ml ray grafana▾
Explainer coming soon — add a paragraph under this heading in CLI.md to replace this placeholder.
Usage: terradev ml ray grafana [OPTIONS]
Access Grafana dashboard.
Options:
--help Show this message and exit.
terradev ml ray install▾
Explainer coming soon — add a paragraph under this heading in CLI.md to replace this placeholder.
Usage: terradev ml ray install [OPTIONS]
Show installation instructions.
Options:
--help Show this message and exit.
terradev ml ray install-monitoring▾
Explainer coming soon — add a paragraph under this heading in CLI.md to replace this placeholder.
Usage: terradev ml ray install-monitoring [OPTIONS]
Install monitoring stack with Ray dashboards.
Options:
--help Show this message and exit.
terradev ml ray list-nodes▾
Explainer coming soon — add a paragraph under this heading in CLI.md to replace this placeholder.
Usage: terradev ml ray list-nodes [OPTIONS]
List cluster nodes.
Options:
--help Show this message and exit.
terradev ml ray metrics-summary▾
Explainer coming soon — add a paragraph under this heading in CLI.md to replace this placeholder.
Usage: terradev ml ray metrics-summary [OPTIONS]
Get comprehensive metrics summary.
Options:
--help Show this message and exit.
terradev ml ray prometheus▾
Explainer coming soon — add a paragraph under this heading in CLI.md to replace this placeholder.
Usage: terradev ml ray prometheus [OPTIONS]
Access Prometheus metrics.
Options:
--help Show this message and exit.
terradev ml ray start▾
Explainer coming soon — add a paragraph under this heading in CLI.md to replace this placeholder.
Usage: terradev ml ray start [OPTIONS]
Start Ray cluster.
Options:
--help Show this message and exit.
terradev ml ray status▾
Explainer coming soon — add a paragraph under this heading in CLI.md to replace this placeholder.
Usage: terradev ml ray status [OPTIONS]
Show cluster status.
Options:
--help Show this message and exit.
terradev ml ray stop▾
Explainer coming soon — add a paragraph under this heading in CLI.md to replace this placeholder.
Usage: terradev ml ray stop [OPTIONS]
Stop Ray cluster.
Options:
--help Show this message and exit.
terradev ml ray test▾
Explainer coming soon — add a paragraph under this heading in CLI.md to replace this placeholder.
Usage: terradev ml ray test [OPTIONS]
Test connection to Ray service.
Options:
--help Show this message and exit.
terradev ml sglang▾
Explainer coming soon — add a paragraph under this heading in CLI.md to replace this placeholder.
Usage: terradev ml sglang [OPTIONS] COMMAND [ARGS]...
SGLang optimization and management with workload-specific auto-tuning
Options:
--help Show this message and exit.
Commands:
detect Auto-detect workload type and show optimization...
install Install SGLang with optimization stack
router Generate cache-aware router command for multi-replica...
sglang-optimize Auto-optimize SGLang configuration for workload type...
start Start optimized SGLang server
test Test SGLang installation and configuration
terradev ml sglang detect▾
Explainer coming soon — add a paragraph under this heading in CLI.md to replace this placeholder.
Usage: terradev ml sglang detect [OPTIONS] MODEL_PATH
Auto-detect workload type and show optimization recommendations
Options:
--workload-type [agentic_chat|batch_inference|low_latency|moe_model|pd_disaggregated|structured_output|rag_workload]
Workload type to test
--user-description TEXT Natural language description of workload
--help Show this message and exit.
terradev ml sglang install▾
Explainer coming soon — add a paragraph under this heading in CLI.md to replace this placeholder.
Usage: terradev ml sglang install [OPTIONS]
Install SGLang with optimization stack
Options:
--instance-ip TEXT Remote instance IP for installation
--ssh-user TEXT SSH user for remote installation
--ssh-key TEXT SSH private key path
--help Show this message and exit.
terradev ml sglang router▾
Explainer coming soon — add a paragraph under this heading in CLI.md to replace this placeholder.
Usage: terradev ml sglang router [OPTIONS] MODEL_PATH
Generate cache-aware router command for multi-replica deployments
Options:
--dp-size INTEGER Data parallel size for multi-replica
--workload-type [agentic_chat|batch_inference|low_latency|moe_model|pd_disaggregated|structured_output|rag_workload]
Workload type for optimization
--help Show this message and exit.
terradev ml sglang sglang-optimize▾
Explainer coming soon — add a paragraph under this heading in CLI.md to replace this placeholder.
Usage: terradev ml sglang sglang-optimize [OPTIONS] MODEL_PATH
Auto-optimize SGLang configuration for workload type and hardware
Options:
--workload-type [agentic_chat|batch_inference|low_latency|moe_model|pd_disaggregated|structured_output|rag_workload]
Workload type for optimization
--user-description TEXT Natural language description of workload
--host TEXT Server host
--port INTEGER Server port
--dry-run Show optimization plan without launching
--help Show this message and exit.
terradev ml sglang start▾
Explainer coming soon — add a paragraph under this heading in CLI.md to replace this placeholder.
Usage: terradev ml sglang start [OPTIONS] MODEL_PATH
Start optimized SGLang server
Options:
--instance-ip TEXT Remote instance IP
--ssh-user TEXT SSH user for remote deployment
--ssh-key TEXT SSH private key path
--workload-type [agentic_chat|batch_inference|low_latency|moe_model|pd_disaggregated|structured_output|rag_workload]
Workload type for optimization
--port INTEGER Server port
--help Show this message and exit.
terradev ml sglang test▾
Explainer coming soon — add a paragraph under this heading in CLI.md to replace this placeholder.
Usage: terradev ml sglang test [OPTIONS]
Test SGLang installation and configuration
Options:
--help Show this message and exit.
terradev ml vllm▾
Explainer coming soon — add a paragraph under this heading in CLI.md to replace this placeholder.
Usage: terradev ml vllm [OPTIONS] COMMAND [ARGS]...
vLLM optimization and management commands.
Options:
--help Show this message and exit.
Commands:
analyze Analyze current vLLM server workload and provide...
auto-optimize Automatically optimize vLLM configuration based on...
benchmark Benchmark vLLM endpoint performance.
optimize Generate optimized vLLM configurations using the 6...
terradev ml vllm analyze▾
Explainer coming soon — add a paragraph under this heading in CLI.md to replace this placeholder.
Usage: terradev ml vllm analyze [OPTIONS]
Analyze current vLLM server workload and provide optimization
recommendations.
Monitors the running vLLM server to understand workload patterns and
generates specific optimization recommendations.
Examples: terradev vllm analyze -e http://localhost:8000 terradev
vllm analyze -e http://10.0.0.1:8000 -d 120
Options:
-e, --endpoint TEXT vLLM endpoint to analyze [required]
-d, --duration INTEGER Analysis duration in seconds
--help Show this message and exit.
terradev ml vllm auto-optimize▾
Explainer coming soon — add a paragraph under this heading in CLI.md to replace this placeholder.
Usage: terradev ml vllm auto-optimize [OPTIONS]
Automatically optimize vLLM configuration based on workload analysis.
Analyzes current workload patterns or sample requests to automatically
select optimal settings for the 6 critical knobs.
Examples: # Analyze running server terradev vllm auto-optimize -e
http://localhost:8000 -m meta-llama/Llama-2-7b-hf
# Analyze from sample file terradev vllm auto-optimize -s
samples.json -m mistralai/Mistral-7B-v0.1 -g 4
# Generate and apply Helm values terradev vllm auto-optimize -e
http://localhost:8000 -m codellama/CodeLlama-34b-hf -o helm
Options:
-e, --endpoint TEXT vLLM endpoint to analyze (if not provided,
uses sample analysis)
-s, --samples PATH JSON file with sample requests
-G, --gpu-count INTEGER Number of GPUs available
-m, --model TEXT Model name [required]
-o, --output [config|args|helm]
Output format
--apply Apply optimizations automatically
--help Show this message and exit.
terradev ml vllm benchmark▾
Explainer coming soon — add a paragraph under this heading in CLI.md to replace this placeholder.
Usage: terradev ml vllm benchmark [OPTIONS]
Benchmark vLLM endpoint performance.
Options:
-e, --endpoint TEXT vLLM endpoint to test [required]
--api-key TEXT vLLM API key
--prompt TEXT Test prompt
-c, --concurrent INTEGER Concurrent requests
--help Show this message and exit.
terradev ml vllm optimize▾
Explainer coming soon — add a paragraph under this heading in CLI.md to replace this placeholder.
Usage: terradev ml vllm optimize [OPTIONS]
Generate optimized vLLM configurations using the 6 critical knobs.
Applies the 6 knobs most teams never touch: 1. --max-num-batched-tokens
(2048→16384 for throughput, 4096 for latency) 2. --gpu-memory-utilization
(0.90→0.95) 3. --max-num-seqs (256/1024→1024 for throughput, 512 for
latency) 4. --enable-prefix-caching (OFF→ON) 5. --enable-chunked-prefill
(OFF→ON) 6. CPU cores (2 + #GPUs for V1 busy loop)
Examples: terradev vllm optimize -m meta-llama/Llama-2-7b-hf -t
throughput terradev vllm optimize -m mistralai/Mistral-7B-v0.1 -t
latency -g 4
Options:
-m, --model TEXT Model name [required]
-t, --type [throughput|latency]
Optimization type
-G, --gpu-count INTEGER Number of GPUs
-o, --output [args|config|helm]
Output format
--help Show this message and exit.
terradev ml wandb▾
Explainer coming soon — add a paragraph under this heading in CLI.md to replace this placeholder.
Usage: terradev ml wandb [OPTIONS] COMMAND [ARGS]...
Weights & Biases experiment tracking with dashboards, reports, and alerts.
Options:
--help Show this message and exit.
Commands:
create-dashboard Create Terradev dashboard in W&B.
create-project Create a new W&B project.
create-report Generate infrastructure report in W&B.
dashboard-status Get comprehensive dashboard status.
list-projects List all W&B projects.
list-runs List recent W&B runs.
setup-alerts Set up Terradev alerts in W&B.
test Test connection to W&B service.
terradev ml wandb create-dashboard▾
Explainer coming soon — add a paragraph under this heading in CLI.md to replace this placeholder.
Usage: terradev ml wandb create-dashboard [OPTIONS]
Create Terradev dashboard in W&B.
Options:
--help Show this message and exit.
terradev ml wandb create-project▾
Explainer coming soon — add a paragraph under this heading in CLI.md to replace this placeholder.
Usage: terradev ml wandb create-project [OPTIONS] PROJECT_NAME
Create a new W&B project.
Options:
--help Show this message and exit.
terradev ml wandb create-report▾
Explainer coming soon — add a paragraph under this heading in CLI.md to replace this placeholder.
Usage: terradev ml wandb create-report [OPTIONS]
Generate infrastructure report in W&B.
Options:
--help Show this message and exit.
terradev ml wandb dashboard-status▾
Explainer coming soon — add a paragraph under this heading in CLI.md to replace this placeholder.
Usage: terradev ml wandb dashboard-status [OPTIONS]
Get comprehensive dashboard status.
Options:
--help Show this message and exit.
terradev ml wandb list-projects▾
Explainer coming soon — add a paragraph under this heading in CLI.md to replace this placeholder.
Usage: terradev ml wandb list-projects [OPTIONS]
List all W&B projects.
Options:
--help Show this message and exit.
terradev ml wandb list-runs▾
Explainer coming soon — add a paragraph under this heading in CLI.md to replace this placeholder.
Usage: terradev ml wandb list-runs [OPTIONS]
List recent W&B runs.
Options:
-l, --limit INTEGER Max runs to return
--help Show this message and exit.
terradev ml wandb setup-alerts▾
Explainer coming soon — add a paragraph under this heading in CLI.md to replace this placeholder.
Usage: terradev ml wandb setup-alerts [OPTIONS]
Set up Terradev alerts in W&B.
Options:
--help Show this message and exit.
terradev ml wandb test▾
Explainer coming soon — add a paragraph under this heading in CLI.md to replace this placeholder.
Usage: terradev ml wandb test [OPTIONS]
Test connection to W&B service.
Options:
--help Show this message and exit.
terradev model-router▾
Explainer coming soon — add a paragraph under this heading in CLI.md to replace this placeholder.
Usage: terradev model-router [OPTIONS] COMMAND [ARGS]...
Model routing cost/quality-aware routing between strong and weak models.
Options:
--help Show this message and exit.
Commands:
classify Classify a message's step type for routing.
configure Configure model routing endpoints and strategy.
llmd-config Generate llm-d KV-cache-aware routing config.
stats Show routing statistics (in-memory, current session).
test Test model routing with a sample prompt.
terradev model-router classify▾
Explainer coming soon — add a paragraph under this heading in CLI.md to replace this placeholder.
Usage: terradev model-router classify [OPTIONS] TEXT
Classify a message's step type for routing.
Options:
--help Show this message and exit.
terradev model-router configure▾
Explainer coming soon — add a paragraph under this heading in CLI.md to replace this placeholder.
Usage: terradev model-router configure [OPTIONS]
Configure model routing endpoints and strategy.
Options:
--strong-url TEXT Strong model endpoint
--strong-model TEXT
--strong-api-key TEXT
--weak-url TEXT Weak model endpoint
--weak-model TEXT
--weak-api-key TEXT Weak model API key (if needed)
--strategy [step_type|threshold|cascade|strong_only|weak_only]
Routing strategy
--cost-threshold FLOAT Complexity threshold for threshold strategy
--help Show this message and exit.
terradev model-router llmd-config▾
Explainer coming soon — add a paragraph under this heading in CLI.md to replace this placeholder.
Usage: terradev model-router llmd-config [OPTIONS]
Generate llm-d KV-cache-aware routing config.
Options:
-f, --format [json|yaml]
--help Show this message and exit.
terradev model-router stats▾
Explainer coming soon — add a paragraph under this heading in CLI.md to replace this placeholder.
Usage: terradev model-router stats [OPTIONS]
Show routing statistics (in-memory, current session).
Options:
-f, --format [json|text]
--help Show this message and exit.
terradev model-router test▾
Explainer coming soon — add a paragraph under this heading in CLI.md to replace this placeholder.
Usage: terradev model-router test [OPTIONS]
Test model routing with a sample prompt.
Options:
-p, --prompt TEXT Test prompt
-f, --format [json|text]
--help Show this message and exit.
terradev monitor▾
Explainer coming soon — add a paragraph under this heading in CLI.md to replace this placeholder.
Usage: terradev monitor [OPTIONS]
Monitor GPU utilization, training metrics, and cost.
Default: nvidia-smi (zero deps). Optional Prometheus/DCGM-exporter hook.
Includes straggler detection for multi-node clusters.
Examples: terradev monitor -n 10.0.0.1 -n 10.0.0.2 -l /tmp/train.log
terradev monitor --from-provision latest --cost-rate 3.50 terradev
monitor --prometheus http://localhost:9090 -f json terradev monitor -j
job-abc123 --interval 5 --count 10
Options:
-j, --job-id TEXT Job ID to monitor
-n, --nodes TEXT Node IPs
--ssh-user TEXT SSH user
--ssh-key TEXT SSH key path
--from-provision TEXT Use nodes from a provision group. "latest" = most
recent.
-l, --log-path TEXT Training log file to parse
-i, --interval FLOAT Snapshot interval in seconds
--count INTEGER Number of snapshots (0 = continuous)
--prometheus TEXT Prometheus endpoint (optional)
--cost-rate FLOAT Cost per GPU-hour in USD
-f, --format [json|text]
--help Show this message and exit.
terradev observe▾
Explainer coming soon — add a paragraph under this heading in CLI.md to replace this placeholder.
Usage: terradev observe [OPTIONS] COMMAND [ARGS]...
Unified observability pipeline - W&B, Phoenix, Cost Analytics with shared
trace ID
Options:
--help Show this message and exit.
Commands:
gateway Observe API Gateway traffic across W&B, Phoenix, and Cost...
status Get status of an observability trace
terradev observe gateway▾
Explainer coming soon — add a paragraph under this heading in CLI.md to replace this placeholder.
Usage: terradev observe gateway [OPTIONS] GATEWAY_ENDPOINT
Observe API Gateway traffic across W&B, Phoenix, and Cost Analytics
Options:
--wandb-project TEXT W&B project name
--wandb-entity TEXT W&B entity/team
--phoenix-endpoint TEXT Phoenix endpoint URL
--enable-cost-analytics / --disable-cost-analytics
Enable cost analytics tracking
--duration INTEGER Observation duration in seconds
--sample-rate FLOAT Sampling rate (0.0-1.0)
--help Show this message and exit.
terradev observe status▾
Explainer coming soon — add a paragraph under this heading in CLI.md to replace this placeholder.
Usage: terradev observe status [OPTIONS] TRACE_ID
Get status of an observability trace
Options:
--help Show this message and exit.
terradev onboarding▾
Explainer coming soon — add a paragraph under this heading in CLI.md to replace this placeholder.
Usage: terradev onboarding [OPTIONS]
Run the interactive onboarding flow
Options:
--force Force onboarding even if already configured
--help Show this message and exit.
terradev optimize▾
Explainer coming soon — add a paragraph under this heading in CLI.md to replace this placeholder.
Usage: terradev optimize [OPTIONS]
Multi-dimensional optimization: cost + performance + kernel optimization
Analyzes running instances for: - Cost optimization (cheaper alternatives) -
CUCo kernel optimization (compute-communication fusion) - Performance tuning
opportunities - Auto-applies optimizations when requested
Options:
--instance-id TEXT Optimize specific instance ID
--auto-apply Automatically apply all recommended optimizations
--help Show this message and exit.
terradev orchestrator▾
Explainer coming soon — add a paragraph under this heading in CLI.md to replace this placeholder.
Usage: terradev orchestrator [OPTIONS] COMMAND [ARGS]...
Model orchestrator for multi-model inference
Options:
--help Show this message and exit.
Commands:
evict Evict a model from GPU memory
infer Test inference with a model
load Load a model into GPU memory
register Register a model with the orchestrator
start Start the model orchestrator for multi-model inference
status Get orchestrator and model status
terradev orchestrator evict▾
Explainer coming soon — add a paragraph under this heading in CLI.md to replace this placeholder.
Usage: terradev orchestrator evict [OPTIONS] MODEL_ID
Evict a model from GPU memory
Options:
--help Show this message and exit.
terradev orchestrator infer▾
Explainer coming soon — add a paragraph under this heading in CLI.md to replace this placeholder.
Usage: terradev orchestrator infer [OPTIONS] MODEL_ID
Test inference with a model
Options:
--help Show this message and exit.
terradev orchestrator load▾
Explainer coming soon — add a paragraph under this heading in CLI.md to replace this placeholder.
Usage: terradev orchestrator load [OPTIONS] MODEL_ID
Load a model into GPU memory
Options:
--force Force loading even if memory is full
--help Show this message and exit.
terradev orchestrator register▾
Explainer coming soon — add a paragraph under this heading in CLI.md to replace this placeholder.
Usage: terradev orchestrator register [OPTIONS] MODEL_ID MODEL_PATH
Register a model with the orchestrator
Options:
--framework [pytorch|vllm|sglang]
Model framework
--priority INTEGER Priority for eviction (higher = less likely
to evict)
--tags TEXT Comma-separated tags for model
categorization
--help Show this message and exit.
terradev orchestrator start▾
Explainer coming soon — add a paragraph under this heading in CLI.md to replace this placeholder.
Usage: terradev orchestrator start [OPTIONS]
Start the model orchestrator for multi-model inference
Options:
--gpu-id INTEGER GPU ID to use for orchestration
--memory-gb FLOAT Total GPU memory in GB
--policy [billing_optimized|latency_optimized|hybrid]
Scaling policy
--help Show this message and exit.
terradev orchestrator status▾
Explainer coming soon — add a paragraph under this heading in CLI.md to replace this placeholder.
Usage: terradev orchestrator status [OPTIONS]
Get orchestrator and model status
Options:
--model-id TEXT Get details for specific model
--help Show this message and exit.
terradev percentiles▾
Explainer coming soon — add a paragraph under this heading in CLI.md to replace this placeholder.
Usage: terradev percentiles [OPTIONS]
Show historical price percentiles (p10p99) per provider.
Options:
-g, --gpu-type TEXT GPU type (e.g. A100, H100) [required]
-p, --provider TEXT Filter to a single provider
--spot Spot instances only
-w, --window INTEGER Lookback window in hours (default: 720 = 30d)
--help Show this message and exit.
terradev preflight▾
Explainer coming soon — add a paragraph under this heading in CLI.md to replace this placeholder.
Usage: terradev preflight [OPTIONS]
Run preflight hardware validation on GPU nodes.
Checks: GPU health (DCGM), NVLink, RDMA, storage I/O, NCCL. All checks run
in parallel via DAGExecutor.
Examples: terradev preflight terradev preflight -n 10.0.0.1 -n
10.0.0.2 --quick terradev preflight --from-provision latest terradev
preflight -f json
Options:
-n, --nodes TEXT Node IPs (multiple allowed, empty = localhost)
--ssh-user TEXT SSH user (default: root)
--ssh-key TEXT SSH key path
--from-provision TEXT Use nodes from a provision group. "latest" = most
recent.
--quick Quick GPU-only check (skip storage/NCCL)
-f, --format [json|text]
--help Show this message and exit.
terradev price-discovery▾
Explainer coming soon — add a paragraph under this heading in CLI.md to replace this placeholder.
Usage: terradev price-discovery [OPTIONS]
Enhanced price discovery with capacity and confidence scoring
Options:
--gpu-type TEXT GPU type for price discovery
--region TEXT Region filter
--hours INTEGER Hours of historical data to analyze
--trends Show price trends
--help Show this message and exit.
terradev providers▾
Explainer coming soon — add a paragraph under this heading in CLI.md to replace this placeholder.
Usage: terradev providers [OPTIONS] COMMAND [ARGS]...
Manage custom provider profiles for intelligent routing
Options:
--help Show this message and exit.
Commands:
export-example Export an example provider profiles YAML file.
list-profiles List all registered provider profiles (built-in and...
load-profiles Load custom provider profiles from a YAML or JSON file.
remove-profile Remove a custom provider profile from the registry.
show-profile Show details for a specific provider profile.
terradev providers export-example▾
Explainer coming soon — add a paragraph under this heading in CLI.md to replace this placeholder.
Usage: terradev providers export-example [OPTIONS]
Export an example provider profiles YAML file.
Example: terradev providers export-example terradev providers export-
example -o ~/.terradev/custom_providers.yaml
Options:
-o, --output PATH Output file path (default: stdout)
--help Show this message and exit.
terradev providers list-profiles▾
Explainer coming soon — add a paragraph under this heading in CLI.md to replace this placeholder.
Usage: terradev providers list-profiles [OPTIONS]
List all registered provider profiles (built-in and custom).
Example: terradev providers list-profiles terradev providers list-
profiles --format json
Options:
-f, --format [table|json|yaml] Output format (default: table)
--help Show this message and exit.
terradev providers load-profiles▾
Explainer coming soon — add a paragraph under this heading in CLI.md to replace this placeholder.
Usage: terradev providers load-profiles [OPTIONS]
Load custom provider profiles from a YAML or JSON file.
Example: terradev providers load-profiles
~/.terradev/custom_providers.yaml terradev providers load-profiles
profiles.json --override
Profile file format (YAML): profiles: my_provider: api_style:
rest auth_type: bearer egress_cost: 0.05 supports_spot:
true
Options:
-p, --path PATH Path to YAML or JSON file containing provider profiles
--override Override existing profiles with same name (default: skip
existing)
--help Show this message and exit.
terradev providers remove-profile▾
Explainer coming soon — add a paragraph under this heading in CLI.md to replace this placeholder.
Usage: terradev providers remove-profile [OPTIONS] NAME
Remove a custom provider profile from the registry.
Example: terradev providers remove-profile my_custom_provider terradev
providers remove-profile my_custom_provider --force
Options:
-f, --force Remove without confirmation
--help Show this message and exit.
terradev providers show-profile▾
Explainer coming soon — add a paragraph under this heading in CLI.md to replace this placeholder.
Usage: terradev providers show-profile [OPTIONS] NAME
Show details for a specific provider profile.
Example: terradev providers show-profile runpod terradev providers show-
profile my_custom_provider --format json
Options:
-f, --format [table|json|yaml] Output format (default: table)
--help Show this message and exit.
terradev provision▾
Explainer coming soon — add a paragraph under this heading in CLI.md to replace this placeholder.
Usage: terradev provision [OPTIONS]
Provision GPU instances across multiple clouds with auto-optimization.
Performs multi-cloud arbitrage: queries all configured providers, builds a
cost-optimized allocation plan, and deploys instances in parallel with
automatic NUMA topology optimization, GPUDirect RDMA, and NCCL tuning.
Examples: terradev provision -g A100 -n 4 # Provision
4x A100 (auto-optimized) terradev provision -g H100 --max-price 2.50
# Provision H100 under $2.50/hr terradev provision -g A100 --dry-run
# Preview plan without launching terradev provision -g RTX4090 --spot
# Force spot instances terradev provision -g A100 --type inference
# Inference workload (auto-selects spot) terradev provision -g H100 -n 8
--parallel 12 # High-throughput training terradev provision -g
RTX4090 --prefer-local # Prefer local GPUs from your pool
Multi-Agent KV Sharing (pass --agents to enable): terradev provision -g
H100 --agents 20 --context 32k --model-name llama-70b terradev provision
-g H100 --agents 50 --context 128k --sharing-topology broadcast --dry-run
terradev provision -g A100 --agents 10 --context 8k --dtype fp8 # fp8
halves KV VRAM
Spot vs On-Demand: - Spot: 60-80% savings, 2-minute termination notice,
auto-checkpointing - On-demand: Guaranteed availability, no interruptions,
higher cost - Auto-selection: Training defaults to on-demand, inference
defaults to spot - Override with --spot or --on-demand flags
Auto-Optimizations (applied automatically): - NUMA alignment: GPU and NIC
on same NUMA node (30-50% bandwidth improvement) - GPUDirect RDMA: Zero-
copy GPU-to-GPU transfers - CPU pinning: Static CPU manager policy -
NCCL tuning: InfiniBand enabled, GDR_LEVEL=PIX
Next Steps: After provisioning: terradev status --live Run commands:
terradev execute -i <instance-id> -c "command" Stop instances: terradev
manage -i <instance-id> -a stop Terminate: terradev manage -i <instance-
id> -a terminate
Options:
-g, --gpu-type TEXT GPU type (required: A100, H100, RTX4090,
L40S, etc.) [required]
-n, --count INTEGER Number of instances to provision (default:
1)
--max-price FLOAT Maximum price per hour in USD (e.g., 2.50)
-p, --providers TEXT Filter to specific providers (multiple
allowed, e.g., runpod,vastai)
--parallel INTEGER Max parallel deploy threads (default: 6)
--dry-run Show allocation plan without launching
instances
--type [training|inference] Workload type (affects spot/on-demand auto-
selection)
--model-name TEXT Model to deploy (for inference workloads)
--endpoint-name TEXT Endpoint name (for inference workloads)
--min-workers INTEGER Minimum workers for auto-scaling (inference)
--max-workers INTEGER Maximum workers for auto-scaling (inference)
--spot Force spot instances (60-80% savings, 2-min
termination notice)
--on-demand Force on-demand instances (guaranteed
availability, higher cost)
--spot-strategy [aggressive|cheapest|balanced|conservative|safe]
Spot instance strategy: aggressive/cheapest,
balanced, conservative/safe (most stable)
--backend [vllm|sglang|dynamo|tensorrt_llm|llmd]
Inference backend: vllm (default), sglang,
dynamo, tensorrt_llm, llmd
--prefer-local Prefer local GPUs from your pool over cloud
providers
--agents INTEGER Number of concurrent agents. Triggers multi-
agent KV VRAM planner.
--context TEXT Context window per agent (e.g. 32k, 128k).
Used with --agents.
--sharing-topology [broadcast|star|chain|none]
KV cache sharing topology between agents
(default: broadcast).
--dtype [fp16|fp8] KV cache dtype. fp8 halves KV VRAM
requirement.
--select TEXT Select instance by number or keyword: 1-N,
cheapest, cheapest-spot, cheapest-secure,
SXM4-40GB, SXM4-80GB, 80GB PCIe
--auto Auto-select cheapest instance without
prompting (CI/CD mode)
--help Show this message and exit.
terradev quote▾
Explainer coming soon — add a paragraph under this heading in CLI.md to replace this placeholder.
Usage: terradev quote [OPTIONS]
Get real-time GPU pricing quotes from all configured providers.
Queries all configured cloud providers in parallel and displays pricing
sorted by cost (cheapest first). Shows spot vs on-demand availability and
estimated monthly costs.
Examples: terradev quote -g A100 # Quote A100 across
all providers terradev quote -g H100 -p runpod,vastai # Quote H100 from
specific providers terradev quote -g RTX4090 -r us-east-1 # Quote
RTX4090 in specific region terradev quote -g A100 -q #
Show quick provision command terradev quote -g RTX4090 --include-local #
Include local GPUs from your pool
Next Steps: After quoting, use: terradev provision -g <gpu-type> Or use
--quick flag to auto-generate provision command
Common GPUs: A100, H100, RTX4090, L40S, V100, L4, T4
Options:
-g, --gpu-type TEXT GPU type to quote (A100, H100, RTX4090, L40S, etc.)
-p, --providers TEXT Filter to specific providers (multiple allowed, e.g.,
runpod,vastai)
--parallel INTEGER Number of parallel queries (default: 6)
-r, --region TEXT Filter by region (e.g., us-east-1, eu-west-1)
-q, --quick Show quick provision command for best quote
--include-local Include local GPUs from your registered pool (priced
at $0/hr)
--help Show this message and exit.
terradev record▾
Explainer coming soon — add a paragraph under this heading in CLI.md to replace this placeholder.
Usage: terradev record [OPTIONS] COMMAND [ARGS]...
Record and export live workflows
Options:
--help Show this message and exit.
Commands:
start Start recording a live workflow
stop Stop recording and optionally export as pipeline
terradev record start▾
Explainer coming soon — add a paragraph under this heading in CLI.md to replace this placeholder.
Usage: terradev record start [OPTIONS]
Start recording a live workflow
Options:
-n, --name TEXT Recording name [required]
--output-dir TEXT Recording output directory
--help Show this message and exit.
terradev record stop▾
Explainer coming soon — add a paragraph under this heading in CLI.md to replace this placeholder.
Usage: terradev record stop [OPTIONS]
Stop recording and optionally export as pipeline
Options:
-n, --name TEXT Recording name [required]
--export TEXT Export as YAML pipeline file
--output-dir TEXT Recording directory
--help Show this message and exit.
terradev reliability▾
Explainer coming soon — add a paragraph under this heading in CLI.md to replace this placeholder.
Usage: terradev reliability [OPTIONS]
Show provider reliability scores and error rates.
Options:
-p, --provider TEXT Filter to a single provider
-w, --window INTEGER Lookback window in hours (default: 720 = 30d)
--ranking Show ranked leaderboard
--help Show this message and exit.
terradev retrain▾
Explainer coming soon — add a paragraph under this heading in CLI.md to replace this placeholder.
Usage: terradev retrain [OPTIONS] COMMAND [ARGS]...
Drift-triggered continuous fine-tuning.
Watch Phoenix traces for quality degradation, auto-retrain LoRA adapters,
evaluate against holdout, and hot-swap onto vLLM zero downtime.
Examples: terradev retrain drift --model llama-70b-prod --source
phoenix-traces terradev retrain status terradev retrain history
Options:
--help Show this message and exit.
Commands:
deploy Manually deploy an adapter from a completed retrain cycle.
detect Check for drift without triggering a retrain.
drift Run a drift-triggered retrain cycle.
history Show retrain cycle history.
terradev retrain deploy▾
Explainer coming soon — add a paragraph under this heading in CLI.md to replace this placeholder.
Usage: terradev retrain deploy [OPTIONS]
Manually deploy an adapter from a completed retrain cycle.
Use this when --auto-swap was not set and eval passed.
Examples: terradev retrain deploy --cycle-id retrain-abc12345 -e
http://10.0.0.1:8000
Options:
--cycle-id TEXT Retrain cycle ID to deploy [required]
-e, --vllm-endpoint TEXT vLLM endpoint [required]
--vllm-api-key TEXT
-f, --format [json|text]
--help Show this message and exit.
terradev retrain detect▾
Explainer coming soon — add a paragraph under this heading in CLI.md to replace this placeholder.
Usage: terradev retrain detect [OPTIONS]
Check for drift without triggering a retrain.
Examples: terradev retrain detect -m llama-70b-prod terradev retrain
detect -m llama-70b-prod --threshold 0.80
Options:
-m, --model TEXT Model identifier [required]
--phoenix-endpoint TEXT
--phoenix-project TEXT
--baseline FLOAT
--threshold FLOAT
--min-samples INTEGER
-f, --format [json|text]
--help Show this message and exit.
terradev retrain drift▾
Explainer coming soon — add a paragraph under this heading in CLI.md to replace this placeholder.
Usage: terradev retrain drift [OPTIONS]
Run a drift-triggered retrain cycle.
Monitors Phoenix traces, detects quality drift, retrains a LoRA adapter,
evaluates it, and optionally hot-swaps it onto a running vLLM server.
Examples: terradev retrain drift -m llama-70b-prod --auto-swap
terradev retrain drift -m llama-70b-prod -e http://10.0.0.1:8000
terradev retrain drift -m llama-70b-prod --eval-threshold 0.90
Options:
-m, --model TEXT Model identifier (e.g. llama-70b-prod)
[required]
--source [phoenix-traces] Data source for drift detection
--method [lora] Fine-tuning method
--eval-threshold FLOAT Minimum eval score to deploy (0.0-1.0)
--deploy [canary|direct] Deployment strategy
--auto-swap Auto-deploy if eval passes (no manual approval)
--phoenix-endpoint TEXT Phoenix collector endpoint
--phoenix-project TEXT Phoenix project name
-e, --vllm-endpoint TEXT vLLM endpoint for eval and deploy
--vllm-api-key TEXT vLLM API key
--baseline FLOAT Baseline quality score
--threshold FLOAT Drift trigger threshold
--min-samples INTEGER Min samples before triggering
-f, --format [json|text]
--help Show this message and exit.
terradev retrain history▾
Explainer coming soon — add a paragraph under this heading in CLI.md to replace this placeholder.
Usage: terradev retrain history [OPTIONS]
Show retrain cycle history.
Examples: terradev retrain history terradev retrain history -n 5 -f
json
Options:
-n, --limit INTEGER Number of cycles to show
-f, --format [json|text]
--help Show this message and exit.
terradev rollback▾
Explainer coming soon — add a paragraph under this heading in CLI.md to replace this placeholder.
Usage: terradev rollback [OPTIONS] JOB_VERSION
EXPLICIT ROLLBACK (versioned manifests)
Options:
--cache-dir TEXT Manifest cache directory
--help Show this message and exit.
terradev run▾
Explainer coming soon — add a paragraph under this heading in CLI.md to replace this placeholder.
Usage: terradev run [OPTIONS]
One-command GPU provisioning, Docker deployment, and workload execution.
Combines provision + deploy + execute into a single step for rapid
prototyping. Automatically selects the cheapest available GPU instance,
pulls the Docker image, configures mounts/ports/env vars, and runs your
workload.
Examples: terradev run -g A100 -i pytorch/pytorch:latest -c "python
train.py" terradev run -g H100 -i vllm/vllm-openai:latest --keep-alive
--port 8000 terradev run -g A100 -i my-training:latest -m
./data:/workspace/data -e WANDB_KEY=xxx terradev run -g RTX4090 -i
ubuntu:latest -c "nvidia-smi" --dry-run
Use Cases: - Quick training runs: terradev run -g A100 -i
pytorch/pytorch:latest -c "python train.py" - Inference serving: terradev
run -g H100 -i vllm/vllm-openai:latest --keep-alive --port 8000 - Data
processing: terradev run -g A100 -i my-image:latest -m ./data:/data -c
"python process.py" - GPU testing: terradev run -g RTX4090 -i
nvidia/cuda:latest -c "nvidia-smi"
Mounts: Format: local_path:container_path Example: -m
./data:/workspace/data -m ./models:/workspace/models
Ports: Expose container ports to access your services Example: --port
8000 (HTTP), --port 22 (SSH)
Environment Variables: Format: KEY=VALUE Example: -e WANDB_KEY=xxx -e
HF_TOKEN=yyy
Keep-Alive Mode: Use --keep-alive for long-running services (inference,
web servers) Instance stays running after command completes Manage with:
terradev manage -i <instance-id> -a stop/terminate
Next Steps: Check status: terradev status --live Run commands: terradev
execute -i <instance-id> -c "command" Stop instance: terradev manage -i
<instance-id> -a stop Terminate: terradev manage -i <instance-id> -a
terminate
Options:
-g, --gpu TEXT GPU type (required: A100, H100, RTX4090, L40S, etc.)
[required]
--image TEXT Docker image (required: e.g., pytorch/pytorch:latest)
[required]
--cmd TEXT Command to run inside the container (e.g., "python
train.py")
-m, --mount TEXT Mount local path:container path (multiple allowed, e.g.,
./data:/workspace/data)
--port INTEGER Ports to expose (multiple allowed, e.g., 8000 for HTTP)
-e, --env TEXT Environment variables KEY=VALUE (multiple allowed, e.g.,
WANDB_KEY=xxx)
--max-price FLOAT Maximum price per hour in USD (e.g., 2.50)
--providers TEXT Filter to specific providers (multiple allowed, e.g.,
runpod,vastai)
--keep-alive Keep instance running after command completes (for
serving)
--dry-run Show deployment plan without executing
--help Show this message and exit.
terradev schedule▾
Explainer coming soon — add a paragraph under this heading in CLI.md to replace this placeholder.
Usage: terradev schedule [OPTIONS] COMMAND [ARGS]...
Spot-aware scheduling for cost-optimized job execution
Options:
--help Show this message and exit.
Commands:
job Schedule a job with spot pricing awareness
list List all scheduled jobs
windows Show available spot pricing windows
terradev schedule job▾
Explainer coming soon — add a paragraph under this heading in CLI.md to replace this placeholder.
Usage: terradev schedule job [OPTIONS] COMMAND GPU_TYPE
Schedule a job with spot pricing awareness
Options:
--cron TEXT Cron expression for recurring jobs
--max-wait-hours INTEGER Maximum hours to wait for optimal pricing
--job-name TEXT Custom job name
--prefer-current / --no-prefer-current
Prefer currently active pricing window
--help Show this message and exit.
terradev schedule list▾
Explainer coming soon — add a paragraph under this heading in CLI.md to replace this placeholder.
Usage: terradev schedule list [OPTIONS]
List all scheduled jobs
Options:
--help Show this message and exit.
terradev schedule windows▾
Explainer coming soon — add a paragraph under this heading in CLI.md to replace this placeholder.
Usage: terradev schedule windows [OPTIONS]
Show available spot pricing windows
Options:
--gpu-type TEXT Filter by GPU type
--help Show this message and exit.
terradev setup▾
Explainer coming soon — add a paragraph under this heading in CLI.md to replace this placeholder.
Usage: terradev setup [OPTIONS] {runpod|vastai|lambda_labs|tensordock|crusoe|b
aseten|coreweave|gcp|aws|azure|oracle}
Get step-by-step setup instructions for any cloud provider.
Shows detailed setup steps including account creation, API key generation,
environment variable configuration, and testing.
Examples: terradev setup runpod # Easiest - 5 minutes terradev
setup vastai # Easiest - 5 minutes terradev setup aws
# Moderate - 30 minutes terradev setup azure --quick # Quick summary
Quick Start: RunPod and Vast.ai are the fastest to set up (5 minutes each)
Use --quick to see just the essential steps
Options:
-q, --quick Show quick setup summary
--help Show this message and exit.
terradev smart-deploy▾
Explainer coming soon — add a paragraph under this heading in CLI.md to replace this placeholder.
Usage: terradev smart-deploy [OPTIONS]
Smart deployment with automatic optimization
Options:
-w, --workload [training|inference|cost-optimized|high-performance]
Workload type (maps to Karpenter
provisioner)
--image TEXT Docker image (e.g. pytorch/pytorch:latest)
[required]
--cmd TEXT Command to run inside the container
-G, --gpu-count INTEGER Number of GPUs (default: per workload
profile)
-b, --budget FLOAT Max $/hr budget forces spot if < $2/hr
-n, --namespace TEXT Kubernetes namespace
--name TEXT Job/Deployment name (auto-generated if
omitted)
-e, --env TEXT Environment variables KEY=VALUE
--mount TEXT Volume mounts host:container
-o, --option INTEGER Deployment option index from smart-deploy
--memory INTEGER Memory in GB
-s, --storage INTEGER Storage in GB
--hours FLOAT Estimated runtime in hours
--region TEXT Preferred region
--dry-run Show recommendation without deploying
--help Show this message and exit.
terradev sso▾
Explainer coming soon — add a paragraph under this heading in CLI.md to replace this placeholder.
Usage: terradev sso [OPTIONS] COMMAND [ARGS]...
Enterprise SSO authentication
Options:
--help Show this message and exit.
Commands:
configure Configure SSO provider
status Show SSO configuration status
test Test SSO provider configuration
terradev sso configure▾
Explainer coming soon — add a paragraph under this heading in CLI.md to replace this placeholder.
Usage: terradev sso configure [OPTIONS]
Configure SSO provider
Options:
-p, --provider [azure_ad|okta|google_workspace|auth0]
SSO provider [required]
--client-id TEXT Client ID (for OIDC providers)
--client-secret TEXT Client secret (for OIDC providers)
--domain TEXT Domain (for Okta/Auth0)
--tenant-id TEXT Tenant ID (for Azure AD)
--entity-id TEXT Entity ID (for SAML providers)
--sso-url TEXT SSO URL (for SAML providers)
--certificate TEXT Certificate (for SAML providers)
--help Show this message and exit.
terradev sso status▾
Explainer coming soon — add a paragraph under this heading in CLI.md to replace this placeholder.
Usage: terradev sso status [OPTIONS]
Show SSO configuration status
Options:
--help Show this message and exit.
terradev sso test▾
Explainer coming soon — add a paragraph under this heading in CLI.md to replace this placeholder.
Usage: terradev sso test [OPTIONS]
Test SSO provider configuration
Options:
-p, --provider TEXT Test specific provider
--help Show this message and exit.
terradev stage▾
Explainer coming soon — add a paragraph under this heading in CLI.md to replace this placeholder.
Usage: terradev stage [OPTIONS]
Compress, chunk, and pre-position datasets near compute.
Supports local files, S3/GCS URIs, HTTP URLs, and HuggingFace dataset names.
Options:
-d, --dataset TEXT Dataset path, S3 URI, GCS URI, HTTP URL, or
HuggingFace name [required]
--target-regions TEXT Comma-separated target regions
--compression [auto|zstd|gzip|none]
Compression algorithm (default: auto)
--plan-only Show staging plan without executing
--help Show this message and exit.
terradev status▾
Explainer coming soon — add a paragraph under this heading in CLI.md to replace this placeholder.
Usage: terradev status [OPTIONS]
Show current status of all provisioned instances and usage statistics.
Displays active GPU instances, their providers, GPU types, pricing, regions,
and current status. Also shows cost analytics and integration status.
Examples: terradev status # Show tracked instances
(fast) terradev status --live # Query providers for live
status (slower) terradev status --format json # Output in JSON
format
Instance Status: - tracked: Instance is tracked in Terradev (may not be
live) - running: Instance is actively running - stopped: Instance is
stopped but allocated - terminated: Instance has been terminated
Next Steps: Manage instances: terradev manage -i <instance-id> -a <action>
Run commands: terradev execute -i <instance-id> -c "command" View costs:
terradev analytics --days 30
Options:
-f, --format [table|json] Output format: table (default) or json
--live Query providers for live instance status (slower
but accurate)
--help Show this message and exit.
terradev train▾
Explainer coming soon — add a paragraph under this heading in CLI.md to replace this placeholder.
Usage: terradev train [OPTIONS] COMMAND [ARGS]...
Launch distributed training jobs across provisioned GPU nodes
Options:
--help Show this message and exit.
Commands:
resume Resume a training job from checkpoint.
start Launch distributed training jobs across provisioned GPU nodes.
status Show training job status, GPU-hours, cost, and ETA.
stop Stop a running training job.
terradev train resume▾
Explainer coming soon — add a paragraph under this heading in CLI.md to replace this placeholder.
Usage: terradev train resume [OPTIONS]
Resume a training job from checkpoint.
Rebuilds config from job state and resumes with topology validation.
Examples: terradev train-resume -j job-abc123 terradev train-resume
-j job-abc123 --checkpoint-id ckpt-xyz
Options:
-j, --job-id TEXT Job ID to resume [required]
--checkpoint-id TEXT Checkpoint to resume from (default: latest)
-f, --format [json|text]
--help Show this message and exit.
terradev train start▾
Explainer coming soon — add a paragraph under this heading in CLI.md to replace this placeholder.
Usage: terradev train start [OPTIONS] [SCRIPT_ARGS]...
Launch distributed training jobs across provisioned GPU nodes.
Orchestrates distributed training with automatic topology optimization,
FlashOptim integration, and checkpoint management. Supports torchrun,
DeepSpeed, Accelerate, and Megatron frameworks.
Examples: terradev train -s train.py --framework torchrun --gpus-per-node
8 terradev train -c job.yaml # Use
YAML config terradev train -s train.py -n 10.0.0.1 -n 10.0.0.2 --tp 2 --
--lr 1e-4 terradev train -s train.py --from-provision latest #
Auto-resolve nodes terradev train -s train.py --from-provision
pg_1709123456_abc12345 terradev train -s train.py --pool workstation-4090
# Use local pool entry terradev train -s train.py --pool workstation-4090
--overflow-to-cloud # Cloud fallback
Workflow: 1. Provision nodes: terradev provision -g H100 -n 4 2.
Validate: terradev preflight (optional, auto-run by default) 3. Train:
terradev train -s train.py --from-provision latest 4. Monitor: terradev
monitor --job <job-id> 5. Checkpoint: terradev checkpoint list --job <job-
id>
FlashOptim (auto-applied): When training with bf16/fp16 and 40GB+ VRAM,
FlashOptim is automatically enabled for gradient compression and
checkpoint optimization.
Frameworks: - torchrun: PyTorch native distributed training (default) -
deepspeed: Microsoft DeepSpeed for large models - accelerate: HuggingFace
Accelerate - megatron: NVIDIA Megatron-LM for massive models
Next Steps: Monitor training: terradev monitor --job <job-id> Check
status: terradev train-status --job <job-id> Stop training: terradev
train-stop --job <job-id> View checkpoints: terradev checkpoint list --job
<job-id>
Options:
-c, --config PATH YAML config file for training job
-s, --script TEXT Training script path (Python file)
--framework [torchrun|deepspeed|accelerate|megatron]
Distributed framework: torchrun (default),
deepspeed, accelerate, megatron
--backend [native|ray] Launch backend: native (default), ray
(optional, requires Ray cluster)
-n, --nodes TEXT Node IP addresses (multiple allowed)
--from-provision TEXT Use nodes from provision group (pg_xxx or
"latest" for most recent)
--pool TEXT Use local pool entry by name (e.g.,
workstation-4090)
--overflow-to-cloud Fall back to cloud providers if local pool
unavailable or insufficient
--gpus-per-node INTEGER GPUs per node (default: 8)
--tp INTEGER Tensor parallel size for model parallelism
--pp INTEGER Pipeline parallel size for model parallelism
--total-steps INTEGER Total training steps (for ETA calculation)
--skip-preflight Skip preflight GPU/NCCL/RDMA validation
checks
-f, --format [json|text] Output format: text (default) or json
--help Show this message and exit.
terradev train status▾
Explainer coming soon — add a paragraph under this heading in CLI.md to replace this placeholder.
Usage: terradev train status [OPTIONS]
Show training job status, GPU-hours, cost, and ETA.
Queries the local SQLite job database no external services needed.
Examples: terradev train-status terradev train-status -j job-abc123
terradev train-status -f json
Options:
-j, --job-id TEXT Job ID (empty = all running)
-f, --format [json|text]
--help Show this message and exit.
terradev train stop▾
Explainer coming soon — add a paragraph under this heading in CLI.md to replace this placeholder.
Usage: terradev train stop [OPTIONS]
Stop a running training job.
Kills training processes on all nodes in parallel.
Examples: terradev train-stop -j job-abc123
Options:
-j, --job-id TEXT Job ID to stop [required]
-f, --format [json|text]
--help Show this message and exit.
terradev triggers▾
Explainer coming soon — add a paragraph under this heading in CLI.md to replace this placeholder.
Usage: terradev triggers [OPTIONS] COMMAND [ARGS]...
Event-driven automation and triggers
Options:
--help Show this message and exit.
Commands:
create Create a new trigger
disable Disable a trigger
enable Enable a trigger
fire Manually fire an event for testing
list List all triggers
terradev triggers create▾
Explainer coming soon — add a paragraph under this heading in CLI.md to replace this placeholder.
Usage: terradev triggers create [OPTIONS] NAME PIPELINE
Create a new trigger
Options:
--type [event|schedule|condition]
Trigger type
--event TEXT Event type to trigger on (dataset_landed,
model_drift_detected, etc.)
--schedule TEXT Cron schedule (e.g., "0 0 * * 0" for Sunday
midnight)
--condition TEXT Condition expression (e.g., "drift_score >
0.1")
--env [dev|staging|prod] Target environment
--help Show this message and exit.
terradev triggers disable▾
Explainer coming soon — add a paragraph under this heading in CLI.md to replace this placeholder.
Usage: terradev triggers disable [OPTIONS] NAME
Disable a trigger
Options:
--help Show this message and exit.
terradev triggers enable▾
Explainer coming soon — add a paragraph under this heading in CLI.md to replace this placeholder.
Usage: terradev triggers enable [OPTIONS] NAME
Enable a trigger
Options:
--help Show this message and exit.
terradev triggers fire▾
Explainer coming soon — add a paragraph under this heading in CLI.md to replace this placeholder.
Usage: terradev triggers fire [OPTIONS] EVENT_TYPE
Manually fire an event for testing
Options:
--data TEXT JSON data for the event
--source TEXT Event source
--help Show this message and exit.
terradev triggers list▾
Explainer coming soon — add a paragraph under this heading in CLI.md to replace this placeholder.
Usage: terradev triggers list [OPTIONS]
List all triggers
Options:
--help Show this message and exit.
terradev up▾
Explainer coming soon — add a paragraph under this heading in CLI.md to replace this placeholder.
Usage: terradev up [OPTIONS]
CLI-native provisioning with manifest cache + drift detection
Options:
-j, --job TEXT Job name for manifest tracking [required]
--cache-dir TEXT Manifest cache directory
--fix-drift Detect and fix drift automatically
--gpu-type TEXT GPU type
--gpu-count INTEGER Number of GPUs
--hours FLOAT Estimated runtime in hours
--budget FLOAT Budget constraint ($/hr)
--region TEXT Preferred region
--dataset TEXT Dataset path for drift detection
--ttl TEXT Time to live for nodes
--help Show this message and exit.
terradev warm-pool▾
Explainer coming soon — add a paragraph under this heading in CLI.md to replace this placeholder.
Usage: terradev warm-pool [OPTIONS] COMMAND [ARGS]...
Warm pool manager for intelligent pre-warming
Options:
--help Show this message and exit.
Commands:
register Register a model with the warm pool manager
start Start the warm pool manager for intelligent pre-warming
status Get warm pool manager status
terradev warm-pool register▾
Explainer coming soon — add a paragraph under this heading in CLI.md to replace this placeholder.
Usage: terradev warm-pool register [OPTIONS] MODEL_ID
Register a model with the warm pool manager
Options:
--priority INTEGER Model priority for warming
--help Show this message and exit.
terradev warm-pool start▾
Explainer coming soon — add a paragraph under this heading in CLI.md to replace this placeholder.
Usage: terradev warm-pool start [OPTIONS]
Start the warm pool manager for intelligent pre-warming
Options:
--strategy [traffic_based|time_based|priority_based|cost_optimized|latency_optimized]
Warm pool strategy
--max-warm INTEGER Maximum models to keep warm
--min-warm INTEGER Minimum models to keep warm
--help Show this message and exit.
terradev warm-pool status▾
Explainer coming soon — add a paragraph under this heading in CLI.md to replace this placeholder.
Usage: terradev warm-pool status [OPTIONS]
Get warm pool manager status
Options:
--help Show this message and exit.