# BestLLMfor

> Independent comparator of locally-runnable LLMs: hardware-matched rankings, benchmark-driven guides, and a public data API. Built and operated by Mohamed Meguedmi.

## Data asset
- [Local LLM Leaderboard](https://bestllmfor.com/leaderboard/): transparent composite ranking of locally-runnable LLMs (CC BY 4.0). Data: https://bestllmfor.com/leaderboard/leaderboard.csv and https://bestllmfor.com/leaderboard/leaderboard.json

## Kits (paid reference guides — one-time payment, updates for life; prices and lifetime bundle on the hub page)
- [The BestLLMfor kits](https://bestllmfor.com/kits/): hub of the 7 kits + the "all the kits, for life" bundle
- [The Local Copilot Kit](https://bestllmfor.com/local-copilot-kit/): replace GitHub Copilot and Cursor with a code assistant that runs 100% on your machine (Ollama + Cline + Aider), 25 chapters + ready-to-paste configs
- [The Local AI Kit](https://bestllmfor.com/local-ai-kit/): your private, offline ChatGPT on your own machine in an hour (LM Studio, Ollama, Open WebUI), non-developers welcome
- [The Local RAG Kit](https://bestllmfor.com/local-rag-kit/): a reliable local RAG over your PDFs, notes and mail, nothing leaves your machine, scripts included
- [The Local Agents Kit](https://bestllmfor.com/local-agents-kit/): agents that act on your machine (agentic Cline, MCP, n8n + Ollama), guardrails included
- [The AI at Work Kit](https://bestllmfor.com/ai-at-work-kit/): deploy on-premise, private AI for a team (privacy law, EU AI Act, multi-user architecture, security, TCO)
- [The Uncensored AI Kit](https://bestllmfor.com/uncensored-ai-kit/): an uncensored model running locally, without abusive refusals on health, history, security or writing, legal framework included
- [The Mac Kit](https://bestllmfor.com/mac-kit/): local LLMs on Apple Silicon (unified memory, MLX vs GGUF, Ollama and LM Studio tuned for your chip)
- [The Local Image AI Kit](https://bestllmfor.com/local-image-ai-kit/): AI images and video generated locally with no subscription — ComfyUI, Flux, Z-Image, Qwen-Image, Wan 2.2; ready-to-load workflows, VRAM tiers, LoRA training, the legal frame (NVIDIA GPU 8-16 GB or Apple Silicon Mac 16 GB+)
- Author of the kits: Mohamed Meguedmi (Falcon Consulting) — https://bestllmfor.com/author/mohamed-meguedmi/

## Measurement provenance

Configurator speeds are heuristic estimates, not measurements on each device. Reproducible experiment and raw data: https://bestllmfor.com/benchmarks/ . French synthetic tasks do not establish English-language quality.