@llm-dev-ops/llm-registry

Command-line interface for the LLM Registry - A secure, production-ready registry for Large Language Models.

Installation

npm install -g @llm-dev-ops/llm-registry

Configuration

Set up your LLM Registry connection:

# Using the config command
llm-registry config --url http://localhost:8080 --token your-api-token

# Or use environment variables
export LLM_REGISTRY_URL=http://localhost:8080
export LLM_REGISTRY_TOKEN=your-api-token

Configuration is saved to ~/.llm-registry.json.

Usage

Health Check

llm-registry health

Model Management

List Models

# List all models
llm-registry models list

# Filter by provider
llm-registry models list --provider openai

# Filter by tags
llm-registry models list --tags gpt,chat

# Limit results
llm-registry models list --limit 20

Get Model Details

llm-registry models get <model-id>

Create a Model

llm-registry models create \
  --name my-model \
  --version 1.0.0 \
  --description "My custom model" \
  --provider openai \
  --tags gpt,chat

Delete a Model

llm-registry models delete <model-id>

Asset Management

List Assets

llm-registry assets list <model-id>

Upload an Asset

llm-registry assets upload <model-id> ./model.safetensors \
  --name weights \
  --version 1.0.0 \
  --content-type application/octet-stream

Download an Asset

llm-registry assets download <model-id> <asset-id> ./output.safetensors

Commands

llm-registry config

Configure the CLI connection settings.

Options:

• -u, --url <url> - Set the base URL
• -t, --token <token> - Set the API token

llm-registry models list

List all models with optional filtering.

Options:

• -p, --provider <provider> - Filter by provider
• -t, --tags <tags> - Filter by tags (comma-separated)
• -l, --limit <limit> - Limit results (default: 10)

llm-registry models get <id>

Get detailed information about a specific model.

llm-registry models create

Create a new model.

Options:

• -n, --name <name> - Model name (required)
• -v, --version <version> - Model version (required)
• -d, --description <description> - Model description
• -p, --provider <provider> - Model provider
• -t, --tags <tags> - Tags (comma-separated)

llm-registry models delete <id>

Delete a model.

llm-registry assets list <model-id>

List all assets for a model.

llm-registry assets upload <model-id> <file>

Upload an asset to a model.

Options:

• -n, --name <name> - Asset name (required)
• -v, --version <version> - Asset version (required)
• -t, --content-type <type> - Content type (default: application/octet-stream)

llm-registry assets download <model-id> <asset-id> <output>

Download an asset from a model.

llm-registry health

Check the health status of the LLM Registry API.

Environment Variables

• LLM_REGISTRY_URL - Base URL of the LLM Registry API
• LLM_REGISTRY_TOKEN - API authentication token

Examples

# Configure the CLI
llm-registry config --url https://registry.example.com

# Create a new model
llm-registry models create \
  --name llama-2-7b \
  --version 1.0.0 \
  --provider meta \
  --tags llama,chat,7b

# Upload model weights
llm-registry assets upload abc123 ./llama-2-7b.safetensors \
  --name weights \
  --version 1.0.0

# List all chat models
llm-registry models list --tags chat

# Download a model
llm-registry assets download abc123 def456 ./downloaded-model.safetensors

# Check API health
llm-registry health

License

Apache-2.0 OR MIT

Contributing

Contributions are welcome! Please see the main repository for guidelines.