ggml-org/ggml
Repository · https://github.com/ggml-org/ggml
Tensor library for machine learning
First seen by Oriel on .
AI summary
ggml is a dependency-free C/C++ tensor library for machine learning, emphasizing portability, efficient execution, broad hardware backends, quantization, and minimal runtime setup.
- The library supports x86, ARM, RISC-V, LoongArch, PowerPC, s390x, and WebAssembly platforms.
- SIMD-optimized kernels target x86, ARM, and RISC-V, while backends support CPU, GPU, NPU, and browser execution.
- ggml provides 2- to 8-bit integer quantization, plus MXFP4 and NVFP4 microscaling formats.
- Runtime execution performs zero memory allocations, and the repository includes a fully commented matrix-multiplication example.
- The project builds with CMake and directs core-library contributors to open pull requests in llama.cpp for visibility and testing.
#ggml #machine learning #tensor library #quantization
Generated by gpt-5.6-sol. Check the original for details.
Metric trends
Last 30 days · Up to 120 recorded samples per source and metric, with reposts kept separate · First and latest retained
Points
Score reported by the source.
- Hacker News: 416, latest observation ; +407 over observed period since
Stars
Total repository stars at each observation.
- GitHub Trending: 16,063, latest observation ; +266 over observed period since
Stars gained (1d)
Daily star gain reported by GitHub Trending. A change compares two daily values; it is not additional stars between observations.
- GitHub Trending: 43, latest observation ; -111 over observed period since
Forks
Total repository forks at each observation.
- GitHub Trending: 2,068, latest observation ; +121 over observed period since
Comments
Comment count reported by the source.
- Hacker News: 68, latest observation ; +68 over observed period since
Aliases
| Source | External ID | URL |
|---|---|---|
| GitHub Trending | ggml-org/ggml | https://github.com/ggml-org/ggml Read discussion |
| Hacker News | 50493062 | https://github.com/ggml-org/ggml/releases/tag/v1.0.0 Read discussion |