mirror of
https://github.com/jart/cosmopolitan.git
synced 2025-01-31 11:37:35 +00:00
e7eb0b3070
- Fix UX issues with llama.com - Do housekeeping on libm code - Add more vectorization to GGML - Get GGJT quantizer programs working well - Have the quantizer keep the output layer as f16c - Prefetching improves performance 15% if you use fewer threads |
||
|---|---|---|
| .. | ||
| c.lang | ||
| cosmo-asm-mode.el | ||
| cosmo-c-builtins.el | ||
| cosmo-c-constants.el | ||
| cosmo-c-keywords.el | ||
| cosmo-c-types.el | ||
| cosmo-cpp-constants.el | ||
| cosmo-format.el | ||
| cosmo-platform-constants.el | ||
| cosmo-stuff.el | ||
| cosmo.el | ||
| ctest-mode.el | ||
| javadown.el | ||
| key.py | ||
| ld-script.el | ||
| optinfo-mode.el | ||
| protobuf-mode.el | ||