Commit history

1. Commits on Sep 15, 2026

   • Aimdo 0.5.5 + Auto-detect and enable --fast-disk when the disk is fast (CORE-440) (#16333)

     rattus128

     authored
     7a0b5eeView commit details

     Copy full SHA for 7a0b5ee

     View code at this pointBrowse repository at this point
2. Commits on Aug 19, 2026

   • Add friendly errors to the dynamic vram safetensors loader. (#15722)

     comfyanonymous

     authored
     a3e5cceView commit details

     Copy full SHA for a3e5cce

     View code at this pointBrowse repository at this point
3. Commits on Jun 22, 2026

   • feat: Support Krea2 (#14589)

     kijai

     authored
     2a61015View commit details

     Copy full SHA for 2a61015

     View code at this pointBrowse repository at this point
4. Commits on May 30, 2026

   • Threaded Loader performance fixes / improvements (+ Aimdo 0.4.6) (#14116)

     Show description for e154da8
     rattus128

     authored
     e154da8View commit details

     Copy full SHA for e154da8

     View code at this pointBrowse repository at this point
5. Commits on May 26, 2026

   • MultiGPU Work Units For Accelerated Sampling (CORE-184) (#7063)

     Kosinkadink

     authored
     0a2dd86View commit details

     Copy full SHA for 0a2dd86

     View code at this pointBrowse repository at this point
6. Commits on May 21, 2026

   • fix(lanczos): correct dimension transposition for single-channel tensors (#12679)

     edoardocarmignani

     authored
     aab41a9View commit details

     Copy full SHA for aab41a9

     View code at this pointBrowse repository at this point

   • Multi-threaded load of models from disk (big load time speedups & Offload to disk) (CORE-43,CORE-152,CORE-164,CORE-165,CORE-117) (#13802)

     Show description for 5aa5ccc
     rattus128

     authored
     5aa5cccView commit details

     Copy full SHA for 5aa5ccc

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

   • Fix void failing with RuntimeError: start (0) + length (464) exceeds dimension size (461). (#13873)

     Talmaj

     authored
     74c17a2View commit details

     Copy full SHA for 74c17a2

     View code at this pointBrowse repository at this point
8. Commits on May 11, 2026

   • chore: remove extra word in comment (#13826)

     box4wangjing

     authored
     f505cb4View commit details

     Copy full SHA for f505cb4

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

   • I don't think there was any because nobody complained. (#13807)

     comfyanonymous

     authored
     66669b2View commit details

     Copy full SHA for 66669b2

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

    • feat: Gemma4 text generation support (CORE-30) (#13376)

      Show description for be95871
      kijai

      authored
      be95871View commit details

      Copy full SHA for be95871

      View code at this pointBrowse repository at this point
11. Commits on Mar 19, 2026

    • Reduce tiled decode peak memory (#13050)

      kijai

      authored
      fd0261dView commit details

      Copy full SHA for fd0261d

      View code at this pointBrowse repository at this point
12. Commits on Mar 17, 2026

    • Fix potential issue. (#13009)

      comfyanonymous

      authored
      ca17fc8View commit details

      Copy full SHA for ca17fc8

      View code at this pointBrowse repository at this point

    • [Trainer] FP4, 8, 16 training by native dtype support and quant linear autograd function (#12681)

      KohakuBlueleaf

      authored
      20561aaView commit details

      Copy full SHA for 20561aa

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

    • comfy aimdo 0.2.11 + Improved RAM Pressure release strategies - Windows speedups (#12925)

      Show description for 7810f49
      rattus128

      authored
      7810f49View commit details

      Copy full SHA for 7810f49

      View code at this pointBrowse repository at this point
14. Commits on Mar 4, 2026

    • DynamicVram: Add casting / fix torch Buffer weights (#12749)

      Show description for ac6513e
      rattus128

      authored
      ac6513eView commit details

      Copy full SHA for ac6513e

      View code at this pointBrowse repository at this point
15. Commits on Feb 25, 2026

    • Fix Aimdo fallback on probe to not use zero-copy sft (#12634)

      Show description for 4f5b7db
      rattus128

      authored
      4f5b7dbView commit details

      Copy full SHA for 4f5b7db

      View code at this pointBrowse repository at this point
16. Commits on Feb 21, 2026

    • comfy-aimdo 0.2 - Improved pytorch allocator integration (#12557)

      Show description for 0bfb936
      rattus128

      authored
      0bfb936View commit details

      Copy full SHA for 0bfb936

      View code at this pointBrowse repository at this point
17. Commits on Feb 19, 2026

    • feat: Add basic text generation support with native models, initially supporting Gemma3 (#12392)

      kijai

      authored
      6d11cc7View commit details

      Copy full SHA for 6d11cc7

      View code at this pointBrowse repository at this point
18. Commits on Feb 15, 2026

    • Remove unsafe pickle loading code that was used on pytorch older than 2.4 (#12473)

      Show description for e1ede29
      comfyanonymous

      authored
      e1ede29View commit details

      Copy full SHA for e1ede29

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

    • Use torch RMSNorm for flux models and refactor hunyuan video code. (#12432)

      comfyanonymous

      authored
      e1add56View commit details

      Copy full SHA for e1add56

      View code at this pointBrowse repository at this point
20. Commits on Feb 11, 2026

    • ace15: Use dynamic_vram friendly trange (#12409)

      Show description for 2a4328d
      rattus128

      authored
      2a4328dView commit details

      Copy full SHA for 2a4328d

      View code at this pointBrowse repository at this point
21. Commits on Feb 10, 2026

    • mp: dont deep-clone objects from model_options (#12382)

      Show description for fe053ba
      rattus128

      authored
      fe053baView commit details

      Copy full SHA for fe053ba

      View code at this pointBrowse repository at this point
22. Commits on Feb 4, 2026

    • utils: safetensors: dont slice data on torch level (#12266)

      Show description for d30c609
      rattus128

      authored
      d30c609View commit details

      Copy full SHA for d30c609

      View code at this pointBrowse repository at this point
23. Commits on Feb 2, 2026

    • dynamic_vram: silence pytorch buffer warning (#12210)

      Show description for 2b5da3b
      rattus128

      authored
      2b5da3bView commit details

      Copy full SHA for 2b5da3b

      View code at this pointBrowse repository at this point
24. Commits on Feb 1, 2026

    • Reduce RAM usage, fix VRAM OOMs, and fix Windows shared memory spilling with adaptive model loading (#11845)

      rattus128

      authored
      f8acd9cView commit details

      Copy full SHA for f8acd9c

      View code at this pointBrowse repository at this point
25. Commits on Jan 21, 2026

    • Add LyCoris LoKr MLP layer support for Flux2 (#11997)

      markuryy

      authored
      0fc1570View commit details

      Copy full SHA for 0fc1570

      View code at this pointBrowse repository at this point
26. Commits on Jan 20, 2026

    • Simpler way to implement the #11980 loras. (#11981)

      comfyanonymous

      authored
      e0eacb0View commit details

      Copy full SHA for e0eacb0

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

    • utils: fix lanczos grayscale upscaling (#11873)

      rattus128

      authored
      80441ebView commit details

      Copy full SHA for 80441eb

      View code at this pointBrowse repository at this point

    • feat: throttle ProgressBar updates to reduce WebSocket flooding (#11504)

      silveroxides

      authored
      712cca3View commit details

      Copy full SHA for 712cca3

      View code at this pointBrowse repository at this point
28. Commits on Jan 5, 2026

    • Support the LTXV 2 model. (#11632)

      comfyanonymous

      authored
      f2b0023View commit details

      Copy full SHA for f2b0023

      View code at this pointBrowse repository at this point
29. Commits on Jan 3, 2026

    • Remove leftover scaled_fp8 key. (#11603)

      comfyanonymous

      authored
      9a552dfView commit details

      Copy full SHA for 9a552df

      View code at this pointBrowse repository at this point
30. Commits on Dec 14, 2025

    • Update warning for old pytorch version. (#11319)

      Show description for 5ac3b26
      comfyanonymous

      authored
      5ac3b26View commit details

      Copy full SHA for 5ac3b26

      View code at this pointBrowse repository at this point
31. Commits on Dec 13, 2025

    • Fix pytorch warnings. (#11314)

      comfyanonymous

      authored
      971cefeView commit details

      Copy full SHA for 971cefe

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

    • dequantization offload accounting (fixes Flux2 OOMs - incl TEs) (#11171)

      Show description for e136b6d
      rattus128

      authored
      e136b6dView commit details

      Copy full SHA for e136b6d

      View code at this pointBrowse repository at this point