Ollama / vLLM Bridge MCP
A community-developed, self-hostable Model Context Protocol server that allows AI agents to manage local Ollama and vLLM model machines and query local LLMs. It supports reading system state, pulling models, switching inference services, and performing LoRA training, ensuring privacy by keeping model traffic local.
Package and access
Local package with source-attributed installation metadata. Koskamo does not execute packages during review.
What you can do with it
- Manage local Ollama and vLLM model machines
- Query local LLMs with schema-validated JSON output
- Pull and remove models from local registries
- Run LoRA training on custom Ollama variants
Source and maintenance
- Last repository push
- Latest release
- v1.0.6
- Maintenance checked
Packages and installation
Source-provided package data. Koskamo has not executed or live-probed this package.
mcp-ollama-vllm 1.0.6
{
"version": "1.0.6",
"transport": "stdio",
"identifier": "mcp-ollama-vllm",
"registryType": "pypi",
"environmentVariables": [
{
"name": "LOCAL_BACKEND",
"format": "string",
"default": "ollama",
"description": "Backend to use: ollama, vllm or openai"
},
{
"name": "LOCAL_HOST",
"format": "string",
"description": "Model endpoint URL; must be local or on your own network. Required for the openai backend"
},
{
"name": "LOCAL_API_KEY",
"format": "string",
"description": "Optional bearer token, openai backend only; sent as a header, never logged"
},
{
"name": "LOCAL_TIMEOUT",
"format": "number",
"default": "600",
"description": "Read timeout in seconds"
},
{
"name": "LOCAL_EMBED_MODEL",
"format": "string",
"default": "nomic-embed-text",
"description": "Default embedding model"
},
{
"name": "LOCAL_EMBED_PATH",
"format": "string",
"default": "/v1/embeddings",
"description": "Optional, openai backend only: embedding endpoint path"
},
{
"name": "LOCAL_SCHEMA_MODE",
"format": "string",
"default": "auto",
"description": "Optional, openai backend only: auto, response_format or none"
}
]
}Tools
Reported by the attributed source; not independently observed by Koskamo.
Maintenance history
Public check and source-refresh events. Editorial and ownership records remain private.
- · repository · passed · metadata changed · updated
- · backfill · recorded · Pre-expansion baseline; unobserved rich metadata remains unknown
Connect from your client
Each snippet configures the same source-provided package; replace placeholders before use.
Cursor
{
"mcpServers": {
"ollama-vllm-bridge-mcp": {
"command": "uvx",
"args": [
"mcp-ollama-vllm==1.0.6"
],
"env": {
"LOCAL_BACKEND": "<YOUR_LOCAL_BACKEND>",
"LOCAL_HOST": "<YOUR_LOCAL_HOST>",
"LOCAL_API_KEY": "<YOUR_LOCAL_API_KEY>",
"LOCAL_TIMEOUT": "<YOUR_LOCAL_TIMEOUT>",
"LOCAL_EMBED_MODEL": "<YOUR_LOCAL_EMBED_MODEL>",
"LOCAL_EMBED_PATH": "<YOUR_LOCAL_EMBED_PATH>",
"LOCAL_SCHEMA_MODE": "<YOUR_LOCAL_SCHEMA_MODE>"
}
}
}
}Cline
{
"mcpServers": {
"ollama-vllm-bridge-mcp": {
"command": "uvx",
"args": [
"mcp-ollama-vllm==1.0.6"
],
"env": {
"LOCAL_BACKEND": "<YOUR_LOCAL_BACKEND>",
"LOCAL_HOST": "<YOUR_LOCAL_HOST>",
"LOCAL_API_KEY": "<YOUR_LOCAL_API_KEY>",
"LOCAL_TIMEOUT": "<YOUR_LOCAL_TIMEOUT>",
"LOCAL_EMBED_MODEL": "<YOUR_LOCAL_EMBED_MODEL>",
"LOCAL_EMBED_PATH": "<YOUR_LOCAL_EMBED_PATH>",
"LOCAL_SCHEMA_MODE": "<YOUR_LOCAL_SCHEMA_MODE>"
}
}
}
}VS Code
{
"servers": {
"ollama-vllm-bridge-mcp": {
"type": "stdio",
"command": "uvx",
"args": [
"mcp-ollama-vllm==1.0.6"
],
"env": {
"LOCAL_BACKEND": "<YOUR_LOCAL_BACKEND>",
"LOCAL_HOST": "<YOUR_LOCAL_HOST>",
"LOCAL_API_KEY": "<YOUR_LOCAL_API_KEY>",
"LOCAL_TIMEOUT": "<YOUR_LOCAL_TIMEOUT>",
"LOCAL_EMBED_MODEL": "<YOUR_LOCAL_EMBED_MODEL>",
"LOCAL_EMBED_PATH": "<YOUR_LOCAL_EMBED_PATH>",
"LOCAL_SCHEMA_MODE": "<YOUR_LOCAL_SCHEMA_MODE>"
}
}
}
}Claude Desktop
{
"mcpServers": {
"ollama-vllm-bridge-mcp": {
"command": "uvx",
"args": [
"mcp-ollama-vllm==1.0.6"
],
"env": {
"LOCAL_BACKEND": "<YOUR_LOCAL_BACKEND>",
"LOCAL_HOST": "<YOUR_LOCAL_HOST>",
"LOCAL_API_KEY": "<YOUR_LOCAL_API_KEY>",
"LOCAL_TIMEOUT": "<YOUR_LOCAL_TIMEOUT>",
"LOCAL_EMBED_MODEL": "<YOUR_LOCAL_EMBED_MODEL>",
"LOCAL_EMBED_PATH": "<YOUR_LOCAL_EMBED_PATH>",
"LOCAL_SCHEMA_MODE": "<YOUR_LOCAL_SCHEMA_MODE>"
}
}
}
}Let your agent find its tools
Find servers like Ollama / vLLM Bridge MCP from your agent. Add Koskamo once to search the index for endpoints, access requirements, and client setup. Public discovery needs no key.