Autoresearch for GPU kernels. Give it any PyTorch model, go to sleep, wake up to optimized Triton kernels.
Repository Index
109,091 visible repositories in the index
1–4 of 4 · Most stars
FP16xINT4 LLM inference kernel that can achieve near-ideal ~4x speedups up to medium batchsizes of 16-32 tokens.
[ICCV 2025] From anything to mesh like human artists. Official impl. of "MeshAnything V2: Artist-Created Mesh Generation With Adjacent Mesh Tokenization"
- 3d
- 3d-generation
- generative-model
- mesh
A Python library which extracts strokes (outlines) from an SVG file as a series of line segments appropriate for driving pen plotters