diff options
| author | Kawrakow <iwankawrakow@gmail.com> | 2025-05-29 18:57:41 +0300 |
|---|---|---|
| committer | GitHub <noreply@github.com> | 2025-05-29 18:57:41 +0300 |
| commit | 1eac9e8487646ee7af00d6d91e10c0cc21ab38c1 (patch) | |
| tree | 63e81bdeaff16aa34bf50aa7d9f4be1a70807391 /gguf-py/scripts/gguf_convert_endian.py | |
| parent | ccd6d9cdf6851f7042c48d682daf47bc0e2eca27 (diff) | |
NEON implementation for trellis quants (#471)
* iq2_kt: NEON implementation
* iq3_kt: NEON implementation
* iq4_kt: not working NEON implementation
* iq4_kt: NEON implementation
Have to use f32 arithmetic else I get gibberish?
Correspondigly ridiculously slow.
* Cleanup
* iq4_kt: slightly faster TG on NEON
---------
Co-authored-by: Iwan Kawrakow <iwan.kawrakow@gmail.com>
Diffstat (limited to 'gguf-py/scripts/gguf_convert_endian.py')
0 files changed, 0 insertions, 0 deletions