QuelLLM provides a powerful Model Context Protocol (MCP) server that grants clients like Claude Code, Cursor, and Continue direct access to the extensive QuelLLM.fr catalog of 250+ open-weight Large Language Models. This server equips users with a suite of tools to efficiently query, compare, and analyze LLMs, including features for estimating VRAM requirements, determining optimal GPU/Mac tiers, and calculating detailed cost comparisons between various API providers and self-hosted hardware setups. All data is sourced from quelllm.fr/api/, ensuring up-to-date information for informed decision-making in LLM deployment.
主な機能
010 GitHub stars
02Detailed LLM model record retrieval
03LLM catalog search and filtering
04Side-by-side LLM model comparison
05VRAM estimation for various quantizations
06API vs. self-hosted LLM cost estimation
