Local AI MCP Server
Run, fit, and route work to local LLMs (Ollama, LM Studio, llama.cpp, MLX, vLLM, and more).
Local AI MCP scores 55 (Fair), with no SPDX license recorded. It installs as a local process. It does not declare tools yet. Filed under AI and memory. Install configs are generated for 10 clients. Being listed in the registry is not the same as being vendor-official.
Fair trust
55 of 100 from six public signals. Nobody can buy a point.
- Listed in the official MCP registry.
- No license recorded.
License, quality and maintenance
Where the 55 comes from
22 + 20 + 0 + 0 + 10 + 3 = 55. Each factor is capped, so no single signal can carry a listing.
Active in the official MCP registry, plus a known vendor namespace.
Registry: 22 of 30 points.How recently the repository saw a commit or release.
Maintenance: 20 of 20 points.GitHub stars and weekly downloads, on a log scale.
Popularity: 0 of 20 points.A clear, OSI-style open-source license.
License: 0 of 10 points.Not archived, a secure install path, a pinned version.
Security: 10 of 10 points.A real summary, a README with depth, listed tools.
Docs: 3 of 10 points.Install
Pick your client, copy the config, paste it in.
Needs Cursor or VS Code installed on this device. The button opens the app and asks you to confirm before it adds the server.
{ "mcpServers": { "tmhsdigital-local-ai-mcp": { "command": "npx", "args": [ "-y", "@tmhs/local-ai-mcp@0.6.1" ], "env": { "OLLAMA_HOST": "http://localhost:11434", "LMSTUDIO_HOST": "http://localhost:1234", "LLAMACPP_HOST": "http://localhost:8080", "DOCKER_MODEL_RUNNER_HOST": "http://localhost:12434", "MLX_HOST": "", "VLLM_HOST": "", "VLLM_API_KEY": "YOUR_VLLM_API_KEY", "FOUNDRY_LOCAL_HOST": "", "OPENAI_COMPAT_HOST": "", "OPENAI_COMPAT_API_KEY": "YOUR_OPENAI_COMPAT_API_KEY", "MOONSHOT_API_KEY": "YOUR_MOONSHOT_API_KEY", "LOCAL_AI_ALLOW_HOSTED_FALLBACK": "", "MOONSHOT_HOST": "https://api.moonshot.ai/v1", "LOCAL_AI_DISABLE": "", "LOCAL_AI_PROVIDERS": "", "LOCAL_AI_DISABLE_NETWORK_SEARCH": "", "LOCAL_AI_BENCHMARK_CACHE": "" } } }}
Environment variables
| Name | Required | Notes |
|---|---|---|
OLLAMA_HOST | Optional | Ollama base URL |
LMSTUDIO_HOST | Optional | LM Studio server base URL |
LLAMACPP_HOST | Optional | llama.cpp server (llama-server) base URL |
DOCKER_MODEL_RUNNER_HOST | Optional | Docker Model Runner host TCP endpoint |
MLX_HOST | Optional | mlx_lm.server base URL (opt-in; Apple Silicon) |
VLLM_HOST | Optional | vLLM OpenAI server base URL (opt-in) |
VLLM_API_KEYSecret | Optional | Bearer token for VLLM_HOST (when vLLM runs with --api-key) |
FOUNDRY_LOCAL_HOST | Optional | Foundry Local service URL (opt-in; port from `foundry service status`) |
OPENAI_COMPAT_HOST | Optional | Any OpenAI-compatible /v1 server (opt-in) |
OPENAI_COMPAT_API_KEYSecret | Optional | Bearer token for OPENAI_COMPAT_HOST |
MOONSHOT_API_KEYSecret | Optional | Moonshot AI (Kimi) key. Hosted: only used when named explicitly. |
LOCAL_AI_ALLOW_HOSTED_FALLBACK | Optional | Allow implicit routing to hosted providers (default: off) |
MOONSHOT_HOST | Optional | Moonshot API base URL |
LOCAL_AI_DISABLE | Optional | Comma-separated provider names to skip probing |
LOCAL_AI_PROVIDERS | Optional | JSON array of extra provider instances: {type, host, name?, apiKey?, kind?} |
LOCAL_AI_DISABLE_NETWORK_SEARCH | Optional | Keep search_models offline (catalog only) |
LOCAL_AI_BENCHMARK_CACHE | Optional | Benchmark results file used to rank delegate by measured speed (default: per-user cache dir; off disables) |
Tools
All capabilitiesTools show up when the registry metadata or the repository README lists them. Public READMEs are read directly. A GitHub token is used only when one is already configured.
About
Run, fit, and route work to local LLMs (Ollama, LM Studio, llama.cpp, MLX, vLLM, and more).
Frequently asked questions
What is Local AI MCP Server?
Run, fit, and route work to local LLMs (Ollama, LM Studio, llama.cpp, MLX, vLLM, and more).
How do I install Local AI MCP Server in Claude Desktop, Cursor, or VS Code?
Pick your client in the Install section on this page and copy the generated config into the client's config file. Step-by-step guides: Claude Desktop, Cursor, and VS Code. Local AI MCP Server runs locally from the npm package @tmhs/local-ai-mcp with the command npx -y @tmhs/local-ai-mcp@0.6.1.
Does Local AI MCP Server support remote (HTTP) connections?
No remote endpoint is listed for Local AI MCP Server. It runs locally on your machine over stdio, started by your MCP client.
How is Local AI MCP Server's trust score calculated?
Local AI MCP Server scores 55 out of 100 (Fair). The score adds six capped public signals: registry status, maintenance, popularity, license, documentation, and security. Nobody can pay for points. Read the methodology for the full formula.
Reviews
No reviews yet—Be the first to review
No published reviews yet. Used it? Tell the next person how it went.
Sign inA verified email is required to review this server.
Add badge to your README
Paste either snippet into the repository README. The image links to this listing.
Manage this listing
Claim this serverRelated servers
mirroir-mcp
by jfarcand · Swift
Drive a stock iPhone via iPhone Mirroring: no Developer Mode, Xcode or WDA. LLM-free skill replay.
Code-Graph-RAG
by vitali87 · Python
Knowledge-graph RAG for multi-language codebases: index, query and edit code in natural language.
ha-mcp
by homeassistant-ai · Python
Configure and automate Home Assistant with Claude, ChatGPT, and other AI assistants.
Mirobody
by thetahealth · Python
Lab results to LOINC codes and UCUM units, symptoms and diagnoses to ICPC-3. Offline, no API key.
projectmem
by riponcm · Python
Coding agent memory — one local MCP server for every project. Warns before repeating failed fixes.
reddit-mcp-buddy
by karanb192 · TypeScript
Browse Reddit RSS; approved API access for search, analysis, and optional posting.