{
  "markdown": "# quelllm-mcp\n\nMCP server exposing the **[quelllm.fr](https://quelllm.fr)** catalog of 190+ open-weights LLMs via Model Context Protocol tools. Use it from Claude Code, Cursor, Continue, or any MCP-compatible client to query models, compare them, estimate VRAM, and compute API vs self-hosted cost.\n\n## Tools exposed\n\n| Tool | Description |\n|---|---|\n| `list_models(filter_origin?, filter_family?, max_params_b?)` | List models with filters (origin code, family, max params in B) |\n| `get_model(model_id)` | Full record for one model (params, vram per quant, context window, family, tags, license, URLs) |\n| `compare(model_a_id, model_b_id)` | Side-by-side comparison with verdict |\n| `estimate_vram(model_id, quant)` | VRAM in GB at chosen quant + recommended GPU/Mac tiers |\n| `estimate_cost(input_tokens_per_month, output_tokens_per_month, ...)` | Cost in EUR — full table API providers vs self-hosted hardware OR a specific id |\n| `search_models(query, limit?)` | Fuzzy search by name, family, tag, author |\n\n## Install\n\nInstall from source (not yet on PyPI) :\n\n```bash\npip install git+https://github.com/MGM-FALCON/quelllm-mcp.git\n```\n\nOr run without installing, using [uv](https://docs.astral.sh/uv/) :\n\n```bash\nuvx --from git+https://github.com/MGM-FALCON/quelllm-mcp.git quelllm-mcp\n```\n\nFor local development :\n\n```bash\ngit clone https://github.com/MGM-FALCON/quelllm-mcp.git\ncd quelllm-mcp\npip install -e .\n```\n\n## Use with Claude Code\n\nAdd to `~/.claude.json` or a project's `.mcp.json`. If you installed with `pip` :\n\n```json\n{\n  \"mcpServers\": {\n    \"quelllm\": {\n      \"command\": \"quelllm-mcp\"\n    }\n  }\n}\n```\n\nOr zero-install with `uvx` :\n\n```json\n{\n  \"mcpServers\": {\n    \"quelllm\": {\n      \"command\": \"uvx\",\n      \"args\": [\"--from\", \"git+https://github.com/MGM-FALCON/quelllm-mcp.git\", \"quelllm-mcp\"]\n    }\n  }\n}\n```\n\n## Use with Claude Desktop\n\nEdit `~/Library/Application Support/Claude/claude_desktop_config.json` (macOS) :\n\n```json\n{\n  \"mcpServers\": {\n    \"quelllm\": {\n      \"command\": \"quelllm-mcp\"\n    }\n  }\n}\n```\n\n## Use with Cursor / Continue / Cline\n\nMost MCP clients accept the same JSON config :\n\n```json\n{\n  \"command\": \"quelllm-mcp\"\n}\n```\n\n## Example queries (from your client)\n\n```\n> Quels LLM Mistral peuvent tourner sur RTX 5070 Ti 16GB ?\n→ list_models(filter_family='Mistral', max_params_b=24)\n→ estimate_vram('mistral-small-24b', 'q4')\n\n> Compare Llama 3.3 70B vs Qwen 2.5 32B\n→ compare('llama33-70b', 'qwen25-32b')\n\n> J'utilise 10M tokens input + 2.5M output / mois. Combien je paye chez OpenAI vs DeepSeek ?\n→ estimate_cost(10_000_000, 2_500_000)\n```\n\n## Data source\n\nAll data pulled from **[quelllm.fr/api/](https://quelllm.fr/api/)** (CC BY 4.0, no key, CORS-enabled). Cached locally for 1h to avoid rate-limiting.\n\nAPI pricing data (GPT-5, Claude Opus 4.7, Gemini 2.5, DeepSeek, Mistral) and hardware pricing (RTX 50-series, Mac M4) are hardcoded as of **2026-05** — verify semestrially.\n\n## License\n\nMIT — see [LICENSE](LICENSE).\n\n## Contributing\n\nSource : https://github.com/MGM-FALCON/quelllm-mcp\nIssues + PRs welcome. Particularly :\n- API pricing updates (semestrial)\n- Hardware additions (new GPUs, Mac Mx series)\n- New tools (e.g. `find_alternatives_to(model_id)`, `recommend_gpu(budget_eur)`)\n\n### Tests\n\nA pytest smoke suite lives under `tests/`. It covers all 6 tools and the v1.1.0\noutput invariants, never touches the network (local fixture + mocked `httpx`),\nand stubs the `mcp` SDK when it isn't importable — so it also runs on Python 3.9.\n\n```bash\npip install -e \".[test]\"\npytest\n```\n\n## Author\n\nMohamed Meguedmi — [LinkedIn](https://linkedin.com/in/mohamed-meguedmi) · [Hugging Face](https://huggingface.co/MGMMMM)\nFounder of [La Gazette IA](https://lagazetteia.fr) and [QuelLLM.fr](https://quelllm.fr).\n",
  "bytes": 3755,
  "sha": "a8d5591557cc3547b75ce3aea5d24065394d00452e095f155ef801f132f932c9",
  "repo_slug": "mgm-falcon/quelllm-mcp",
  "fonte": "repo",
  "truncated": false,
  "api": "https://agentalog.com/api/listings/mcp_io_github_mgm_falcon_quelllm_mcp_54fef581/readme"
}