llmfit scans your hardware and tells you exactly which local LLMs will actually run

thisdudelikesAI · x · 2026-11-16

llmfit is an open-source CLI that scans your RAM, CPU, GPU and VRAM, then scores hundreds of models (Meta, Mistral, Qwen, DeepSeek and more) across four dimensions: quality (parameter count and quantization), estimated tokens/sec for your exact backend, memory fit, and context window support. Each model gets a Perfect/Good/Marginal/Too Tight label, the tool auto-picks the best quantization by stepping down until it fits, and it works out of the box with Ollama, llama.cpp, MLX and LM Studio.

Original post →

More from Infra

Infra channel →