Commit history

1. Commits on Sep 24, 2025

   • Context Parallel w/ Ring & Ulysses & Unified Attention (#11941)

     Show description for dcb6dd9
     

     3 peopleauthored
     dcb6dd9View commit details

     Copy full SHA for dcb6dd9

     View code at this pointBrowse repository at this point

   • Fix WanVACEPipeline to allow prompt to be None and skip encoding step (#12251)

     Show description for 043ab25
     albchim

     authored
     043ab25View commit details

     Copy full SHA for 043ab25

     View code at this pointBrowse repository at this point

   • xpu enabling for 4 cases (#12345)

     Show description for 7a58734
     yao-matrix

     authored
     7a58734View commit details

     Copy full SHA for 7a58734

     View code at this pointBrowse repository at this point

   • Fix Custom Code loading (#12378)

     Show description for 7c54a7b
     DN6

     authored
     7c54a7bView commit details

     Copy full SHA for 7c54a7b

     View code at this pointBrowse repository at this point
2. Commits on Sep 23, 2025

   • Allow Automodel to support custom model code (#12353)

     Show description for 80de641
     DN6

     authored
     80de641View commit details

     Copy full SHA for 80de641

     View code at this pointBrowse repository at this point
3. Commits on Sep 22, 2025

   • [Fix] chroma docs (#12360)

     Show description for 1448b03
     SahilCarterr

     authored
     1448b03View commit details

     Copy full SHA for 1448b03

     View code at this pointBrowse repository at this point

   • [lora] factor out the overlaps in `save_lora_weights()`. (#12027)

     Show description for d8310a8
     sayakpaul

     authored
     d8310a8View commit details

     Copy full SHA for d8310a8

     View code at this pointBrowse repository at this point

   • [Fix] enable_xformers_memory_efficient_attention() in Flux Pipeline (#12337)

     Show description for 78031c2
     SahilCarterr

     authored
     78031c2View commit details

     Copy full SHA for 78031c2

     View code at this pointBrowse repository at this point

   • Fix bug with VAE slicing in autoencoder_dc.py (#12343)

     mingyi456

     authored
     d83d35cView commit details

     Copy full SHA for d83d35c

     View code at this pointBrowse repository at this point

   • [tests] xfail some kandinsky tests. (#12364)

     Show description for 843355f
     sayakpaul

     authored
     843355fView commit details

     Copy full SHA for 843355f

     View code at this pointBrowse repository at this point
4. Commits on Sep 21, 2025

   • feat: Add QwenImageEditPlus to support future feature upgrades (#12357)

     Show description for df267ee
     naykun

     authored
     df267eeView commit details

     Copy full SHA for df267ee

     View code at this pointBrowse repository at this point
5. Commits on Sep 20, 2025

   • [CI] Fix TRANSFORMERS_FLAX_WEIGHTS_NAME import issue (#12354)

     Show description for edd614e
     DN6

     authored
     edd614eView commit details

     Copy full SHA for edd614e

     View code at this pointBrowse repository at this point
6. Commits on Sep 18, 2025

   • Convert alphas for embedders for sd-scripts to ai toolkit conversion (#12332)

     Show description for 7e7e62c
     

     3 peopleauthored
     7e7e62cView commit details

     Copy full SHA for 7e7e62c

     View code at this pointBrowse repository at this point
7. Commits on Sep 17, 2025

   • Fix many type hint errors (#12289)

     Show description for efb7a29
     DefTruth

     authored
     efb7a29View commit details

     Copy full SHA for efb7a29

     View code at this pointBrowse repository at this point
8. Commits on Sep 16, 2025

   • Fix autoencoder_kl_wan.py bugs for Wan2.2 VAE (#12335)

     Show description for d06750a
     franciszzj

     authored
     d06750aView commit details

     Copy full SHA for d06750a

     View code at this pointBrowse repository at this point

   • Added LucyEditPipeline (#12340)

     Show description for 8c72cd1
     

     sarihl

     and

     yiyixuxu

     authored
     8c72cd1View commit details

     Copy full SHA for 8c72cd1

     View code at this pointBrowse repository at this point

   • fixed bug in defining embed dim for UNet1D (#12111)

     Show description for 751e250
     

     SammyAgrawal

     and

     DN6

     authored
     751e250View commit details

     Copy full SHA for 751e250

     View code at this pointBrowse repository at this point
9. Commits on Sep 15, 2025

   • Add Wan2.2 VACE - Fun (#12324)

     Show description for b500140
     

     linoytsaban

     and

     github-actions[bot]

     authored
     b500140View commit details

     Copy full SHA for b500140

     View code at this pointBrowse repository at this point
10. Commits on Sep 12, 2025

    • Use SDP on BF16 in GPU/HPU migration (#12310)

      Show description for f5c113e
      dsocek

      authored
      f5c113eView commit details

      Copy full SHA for f5c113e

      View code at this pointBrowse repository at this point
11. Commits on Sep 11, 2025

    • Deprecate slicing and tiling methods from `DiffusionPipeline` (#12271)

      Show description for 5e181ed
      sayakpaul

      authored
      5e181edView commit details

      Copy full SHA for 5e181ed

      View code at this pointBrowse repository at this point
12. Commits on Sep 10, 2025

    • Fix AttributeError of `VisualClozeProcessor` (#12121)

      Show description for 55f0b3d
      

      Justin900429

      and

      yiyixuxu

      authored
      55f0b3dView commit details

      Copy full SHA for 55f0b3d

      View code at this pointBrowse repository at this point

    • [quant] allow `components_to_quantize` to be a non-list for single components (#12234)

      Show description for eb7ef26
      

      sayakpaul

      and

      stevhliu

      authored
      eb7ef26View commit details

      Copy full SHA for eb7ef26

      View code at this pointBrowse repository at this point

    • fix for the qwen controlnet pipeline - wrong device can be used (#12309)

      Show description for e1b7f1f
      ttio2tech

      authored
      e1b7f1fView commit details

      Copy full SHA for e1b7f1f

      View code at this pointBrowse repository at this point

    • [feat] cache allocator warmup for `from_single_model` (#12305)

      Show description for 9e7ae56
      sayakpaul

      authored
      9e7ae56View commit details

      Copy full SHA for 9e7ae56

      View code at this pointBrowse repository at this point

    • [modular] fix flux modular pipelines for t2i and i2i (#12272)

      Show description for f7b7945
      sayakpaul

      authored
      f7b7945View commit details

      Copy full SHA for f7b7945

      View code at this pointBrowse repository at this point

    • [core] feat: support group offloading at the pipeline level (#12283)

      Show description for 4345907
      

      sayakpaul

      and

      stevhliu

      authored
      4345907View commit details

      Copy full SHA for 4345907

      View code at this pointBrowse repository at this point
13. Commits on Sep 9, 2025

    • gguf new quant type support (with demo) (#12076)

      Show description for 28106fc
      

      calcuis

      and

      github-actions[bot]

      authored
      28106fcView commit details

      Copy full SHA for 28106fc

      View code at this pointBrowse repository at this point

    • Support ControlNet-Inpainting for Qwen-Image (#12301)

      Show description for 4e36bb0
      

      3 peopleauthored
      4e36bb0View commit details

      Copy full SHA for 4e36bb0

      View code at this pointBrowse repository at this point
14. Commits on Sep 8, 2025

    • [Modular] Qwen (#12220)

      Show description for f50b18e
      yiyixuxu

      authored
      f50b18eView commit details

      Copy full SHA for f50b18e

      View code at this pointBrowse repository at this point
15. Commits on Sep 3, 2025

    • fix some typos (#12265)

      Show description for 764b624
      co63oc

      authored
      764b624View commit details

      Copy full SHA for 764b624

      View code at this pointBrowse repository at this point

    • Add AttentionMixin to WanVACETransformer3DModel (#12268)

      Show description for 6682956
      J4BEZ

      authored
      6682956View commit details

      Copy full SHA for 6682956

      View code at this pointBrowse repository at this point

    • [Quantization] Add TRT-ModelOpt as a Backend (#11173)

      Show description for 4acbfbf
      

      ishan-modi

      and

      sayakpaul

      authored
      4acbfbfView commit details

      Copy full SHA for 4acbfbf

      View code at this pointBrowse repository at this point

    • [core] use `kernels` to support `_flash_3_hub` attention backend (#12236)

      Show description for 130fd8d
      sayakpaul

      authored
      130fd8dView commit details

      Copy full SHA for 130fd8d

      View code at this pointBrowse repository at this point
16. Commits on Sep 1, 2025

    • Fix lora conversion function for ai-toolkit Qwen Image LoRAs (#12261)

      Show description for 901da9d
      

      apolinario

      and

      sayakpaul

      authored
      901da9dView commit details

      Copy full SHA for 901da9d

      View code at this pointBrowse repository at this point
17. Commits on Aug 31, 2025

    • Add Qwen-Image-Edit Inpainting pipeline (#12225)

      Show description for 67ffa70
      

      4 peopleauthored
      67ffa70View commit details

      Copy full SHA for 67ffa70

      View code at this pointBrowse repository at this point