Repository navigation

[OMNIML-5899] Add Q8_0 quantization codec and backend - #2516

Merged

hychiang-git merged 9 commits into

mainfrom

hungyuehc/q8-0-quantization

Oct 2, 2026

Merged

[OMNIML-5899] Add Q8_0 quantization codec and backend#2516
hychiang-git merged 9 commits into
mainfrom
hungyuehc/q8-0-quantization

Commits on Sep 24, 2026
- [OMNIML-5899] Add Q8_0 quantization codec
  hychiang-git
  committed
  489ca8a
Commits on Sep 28, 2026
- Merge branch 'main' into hungyuehc/q8-0-quantization
  hychiang-git
  authored
  14df57e
- [OMNIML-5899] Address Q8_0 review feedback
  hychiang-git
  committed
  0b5461f
- Merge branch 'main' into hungyuehc/q8-0-quantization
  hychiang-git
  authored
  e0eab44
- fix: align Q8_0 CUDA rounding with reference
  hychiang-git
  committed
  6d43a4c
Commits on Sep 30, 2026
- Add the IQ1_M CUDA encoder and register the format
  
  cjluo-nv
  and
  claude
  committed
  49f8d1e
- Merge latest main and IQ1_M registration into Q8_0 codec
  hychiang-git
  committed
  3c0cfd1
Commits on Oct 1, 2026
- Merge main and resolve the GGML format registry conflict
  hychiang-git
  committed
  2beb70f
Commits on Oct 2, 2026
- Merge branch 'main' into hungyuehc/q8-0-quantization
  hychiang-git
  authored
  71e3183