Show HN: What LLM fits my GPU? – VRAM measured on real hardware for 117 models

Category: ai-ml

Tags: llm-inference, quantization, cpp, gpu-acceleration, vram

Score: 8.5/10 (Innovation: 7, Technical: 9, Documentation: 9, Utility: 9)

llama.cpp is a C/C++ implementation for LLM inference with minimal dependencies, supporting a wide range of hardware backends and quantization methods. The related landing page quantized.uk provides a VRAM index for 117 models measured on real hardware, helping users choose models that fit their GPU. The project is highly influential, enabling efficient local and cloud-based LLM deployment.

Target audience: AI researchers, machine learning engineers, backend developers, devops

Repository: https://quantized.uk/ · C++ · MIT · 132067 stars

View on Hacker News