Commit history

1. Commits on Sep 29, 2026

   • Support w6a8 quantization format (#16483)

     kijai

     authored
     491bf3bView commit details

     Copy full SHA for 491bf3b

     View code at this pointBrowse repository at this point
2. Commits on Sep 9, 2026

   • quant_ops: drop the dead ROCm triton arch gate (#16211)

     0xDELUXA

     authored
     be92396View commit details

     Copy full SHA for be92396

     View code at this pointBrowse repository at this point
3. Commits on Aug 11, 2026

   • Make cu130 warning more visible. (#15463)

     comfyanonymous

     authored
     4f3544dView commit details

     Copy full SHA for 4f3544d

     View code at this pointBrowse repository at this point
4. Commits on Aug 7, 2026

   • Support asym w4a8_int (#15308)

     Show description for 344b439
     kijai

     authored
     344b439View commit details

     Copy full SHA for 344b439

     View code at this pointBrowse repository at this point
5. Commits on Aug 3, 2026

   • feat: Support MiniMax-H3 (CORE-375) (#15224)

     kijai

     authored
     57500fcView commit details

     Copy full SHA for 57500fc

     View code at this pointBrowse repository at this point
6. Commits on Jul 11, 2026

   • Temporarily disable auto enabling triton by default on AMD. (#14878)

     Show description for f3a36e7
     comfyanonymous

     authored
     f3a36e7View commit details

     Copy full SHA for f3a36e7

     View code at this pointBrowse repository at this point
7. Commits on Jul 10, 2026

   • Only auto-enable the ROCm comfy-kitchen Triton backend on matrix-core GPUs (#14869)

     Show description for 1377a2f
     liminfei-amd

     authored
     1377a2fView commit details

     Copy full SHA for 1377a2f

     View code at this pointBrowse repository at this point

   • Enable comfy-kitchen Triton backend by default on ROCm/AMD (#14862)

     Show description for 099522f
     liminfei-amd

     authored
     099522fView commit details

     Copy full SHA for 099522f

     View code at this pointBrowse repository at this point
8. Commits on Jul 9, 2026

   • Support convrot int4 models. (#14859)

     Show description for 73e84d5
     comfyanonymous

     authored
     73e84d5View commit details

     Copy full SHA for 73e84d5

     View code at this pointBrowse repository at this point
9. Commits on Jun 25, 2026

   • Support int8 models. (#14636)

     comfyanonymous

     authored
     1a510f0View commit details

     Copy full SHA for 1a510f0

     View code at this pointBrowse repository at this point
10. Commits on May 3, 2026

    • Enable triton comfy kitchen via cli-arg (#12730)

      silveroxides

      authored
      b138133View commit details

      Copy full SHA for b138133

      View code at this pointBrowse repository at this point
11. Commits on Mar 14, 2026

    • feat: Support mxfp8 (#12907)

      kijai

      authored
      1c5db73View commit details

      Copy full SHA for 1c5db73

      View code at this pointBrowse repository at this point
12. Commits on Jan 14, 2026

    • Optimize nvfp4 lora applying. (#11866)

      Show description for 6165c38
      comfyanonymous

      authored
      6165c38View commit details

      Copy full SHA for 6165c38

      View code at this pointBrowse repository at this point
13. Commits on Jan 13, 2026

    • Make loras work on nvfp4 models. (#11837)

      Show description for b3c0e4d
      comfyanonymous

      authored
      b3c0e4dView commit details

      Copy full SHA for b3c0e4d

      View code at this pointBrowse repository at this point
14. Commits on Jan 8, 2026

    • Add warning for old pytorch. (#11718)

      comfyanonymous

      authored
      21e8425View commit details

      Copy full SHA for 21e8425

      View code at this pointBrowse repository at this point
15. Commits on Jan 7, 2026

    • Disable comfy kitchen cuda if pytorch cuda less than 13 (#11681)

      comfyanonymous

      authored
      edee33fView commit details

      Copy full SHA for edee33f

      View code at this pointBrowse repository at this point
16. Commits on Jan 6, 2026

    • Initial ops changes to use comfy_kitchen: Initial nvfp4 checkpoint support. (#11635)

      Show description for 6da00dd
      

      comfyanonymous

      and

      Kosinkadink

      authored
      6da00ddView commit details

      Copy full SHA for 6da00dd

      View code at this pointBrowse repository at this point
17. Commits on Dec 9, 2025

    • Fix nan issue when quantizing fp16 tensor. (#11213)

      comfyanonymous

      authored
      791e30fView commit details

      Copy full SHA for 791e30f

      View code at this pointBrowse repository at this point
18. Commits on Dec 5, 2025

    • Make old scaled fp8 format use the new mixed quant ops system. (#11000)

      comfyanonymous

      authored
      43071e3View commit details

      Copy full SHA for 43071e3

      View code at this pointBrowse repository at this point
19. Commits on Nov 28, 2025

    • fix QuantizedTensor.is_contiguous (#10956) (#10959)

      urlesistiana

      authored
      6484ac8View commit details

      Copy full SHA for 6484ac8

      View code at this pointBrowse repository at this point
20. Commits on Nov 27, 2025

    • quant ops: Dequantize weight in-place (#10935)

      Show description for 3f382a4
      rattus128

      authored
      3f382a4View commit details

      Copy full SHA for 3f382a4

      View code at this pointBrowse repository at this point
21. Commits on Nov 26, 2025

    • Fix loras not working on mixed fp8. (#10899)

      comfyanonymous

      authored
      bdb10a5View commit details

      Copy full SHA for bdb10a5

      View code at this pointBrowse repository at this point
22. Commits on Nov 25, 2025

    • I found a case where this is needed (#10875)

      comfyanonymous

      authored
      015a059View commit details

      Copy full SHA for 015a059

      View code at this pointBrowse repository at this point

    • Allow pinning quantized tensors. (#10873)

      comfyanonymous

      authored
      b680542View commit details

      Copy full SHA for b680542

      View code at this pointBrowse repository at this point

    • Cleanup and fix issues with text encoder quants. (#10872)

      comfyanonymous

      authored
      25022e0View commit details

      Copy full SHA for 25022e0

      View code at this pointBrowse repository at this point
23. Commits on Nov 12, 2025

    • Quantized Ops fixes (#10715)

      Show description for 3b3ef9a
      contentis

      authored
      3b3ef9aView commit details

      Copy full SHA for 3b3ef9a

      View code at this pointBrowse repository at this point
24. Commits on Nov 4, 2025

    • More fp8 torch.compile regressions fixed. (#10625)

      comfyanonymous

      authored
      af4b7b5View commit details

      Copy full SHA for af4b7b5

      View code at this pointBrowse repository at this point

    • Bring back fp8 torch compile performance to what it should be. (#10622)

      comfyanonymous

      authored
      6b88478View commit details

      Copy full SHA for 6b88478

      View code at this pointBrowse repository at this point
25. Commits on Nov 3, 2025

    • Fixes (#10621)

      comfyanonymous

      authored
      e199c8cView commit details

      Copy full SHA for e199c8c

      View code at this pointBrowse repository at this point

    • People should update their pytorch versions. (#10618)

      comfyanonymous

      authored
      958a171View commit details

      Copy full SHA for 958a171

      View code at this pointBrowse repository at this point
26. Commits on Nov 1, 2025

    • Fix torch compile regression on fp8 ops. (#10580)

      comfyanonymous

      authored
      c58c13bView commit details

      Copy full SHA for c58c13b

      View code at this pointBrowse repository at this point
27. Commits on Oct 29, 2025

    • Fix small performance regression with fp8 fast and scaled fp8. (#10537)

      comfyanonymous

      authored
      906c089View commit details

      Copy full SHA for 906c089

      View code at this pointBrowse repository at this point

    • Reduce memory usage for fp8 scaled op. (#10531)

      comfyanonymous

      authored
      1a58087View commit details

      Copy full SHA for 1a58087

      View code at this pointBrowse repository at this point
28. Commits on Oct 28, 2025

    • Mixed Precision Quantization System (#10498)

      Show description for 8817f8f
      contentis

      authored
      8817f8fView commit details

      Copy full SHA for 8817f8f

      View code at this pointBrowse repository at this point
29. Loading