Mohamed Meguedmi
Independent engineer & LLM researcher. Every guide, ranking and comparison on BestLLMfor is researched, benchmarked and written by him.
Who writes BestLLMfor
Mohamed Meguedmi is a France-based independent engineer who runs LLMs on his own hardware. BestLLMfor is his English-language project: an independent comparator of locally-runnable large language models, built on a benchmark pipeline he operates himself and on structured data published as an open data API (CC BY 4.0).
He focuses on what actually matters when you self-host a model: VRAM budgets, quantization trade-offs, real tokens-per-second on consumer GPUs and Apple Silicon, licensing, and total cost versus cloud APIs.
Why you can check his work
- Methodology — how models are tested, scored and ranked (open and reproducible).
- About BestLLMfor — why this site exists, how the data is built, and the no-affiliate policy.
- Benchmarks — the measured numbers behind the recommendations.
- Open data API — the full catalog, free to reuse under CC BY 4.0.
Read his work
Start with the guides, the hardware and use-case rankings, or the head-to-head comparisons.