Get the latest tech news
Promising results from DeepSeek R1 for code
PR by Xuan-Son Nguyen for `llama.cpp`: > This PR provides a big jump in speed for WASM by leveraging SIMD instructions for `qX_K_q8_K` and `qX_0_q8_0` dot product functions. > > …
This PR provides a big jump in speed for WASM by leveraging SIMD instructions for qX_K_q8_K and qX_0_q8_0 dot product functions. The only thing I do is to develop tests and write prompts (with some trails and errors) (It thought about model_map a lot before finally deciding to eliminate it, which was also my preferred resolution.)
Or read this on Hacker News
