search
CUDA
Trends
- 1
Salvatore Sanfilippo, the creator of Redis known as antirez, has released ds4, a local inference engine for DeepSeek 4 Flash and PRO models. The project, written in C, supports Metal, CUDA and ROCm, meaning it runs on Apple, Nvidia and AMD hardware. It is drawing attention as a lightweight option for running the Chinese models entirely on local machines.
Repos
- antirez/ds4 DeepSeek 4 Flash and PRO local inference engine for Metal, CUDA and ROCm
- tile-ai/tilelang Domain-specific language designed to streamline the development of high-performance GPU/CPU/Accelerators kernels
- magnitudedev/magnitude Open source inference engine for agents that optimizes itself for your exact hardware. Compiles and tunes its kernels on
- tensorflow/tensorflow An Open Source Machine Learning Framework for Everyone
- lostmsu/TurboGPT Train a tiny GPT in under a minute (CUDA only)