Commits

Branch selector

 master

User selector

All users

Datepicker

All time

Commit History

Commits on Oct 30, 2024

• Support SD3.5 medium diffusers format weights and loras.

  comfyanonymous

  committed
  09fdb2b

Commits on Oct 29, 2024

• Update web content to release v1.3.26 (comfyanonymous#5413)

  Show description for 65a8659
  huchenlei

  authored
  65a8659

• Cleanup SkipLayerGuidanceSD3 node.

  comfyanonymous

  committed
  770ab20

• SLG first implementation for SD3.5 (comfyanonymous#5404)

  Show description for 954683d
  Dango233

  authored
  954683d

• Add a way to patch blocks in SD3.

  comfyanonymous

  committed
  30c0c81

• Update SD3 code.

  comfyanonymous

  committed
  13b0ff8

Commits on Oct 28, 2024

• Remove useless line.

  comfyanonymous

  committed
  c320801

• Update web content to release v1.3.21 (comfyanonymous#5351)

  Show description for c0b0cfa
  huchenlei

  authored
  c0b0cfa

• Set default shift on mochi to 6.0

  comfyanonymous

  committed
  669d9e4

Commits on Oct 27, 2024

• float16 inference is a bit broken on mochi.

  comfyanonymous

  committed
  9ee0a65

Commits on Oct 26, 2024

• Basic Genmo Mochi video model support.

  Show description for 5cbb01b
  comfyanonymous

  committed
  5cbb01b

• Make LatentUpscale nodes work on 3d latents.

  comfyanonymous

  committed
  c3ffbae

Commits on Oct 25, 2024

• Make euler_ancestral work on flow models (credit: Ashen).

  comfyanonymous

  committed
  d605677

• Revert download to .tmp in frontend_management (comfyanonymous#5369)

  huchenlei

  authored
  ce759b7

Commits on Oct 24, 2024

• Add a model merge node for SD3.5 large.

  comfyanonymous

  committed
  5281090

• support SimpleTuner lycoris lora for SD3 (comfyanonymous#5340)

  PsychoLogicAu

  authored
  af8cf79

Commits on Oct 23, 2024

• Fix ControlLora issue with last commit.

  comfyanonymous

  committed
  66b0961

• Clean up some controlnet code.

  Show description for 754597c
  comfyanonymous

  committed
  754597c

Commits on Oct 22, 2024

• Fix lowvram edge case.

  comfyanonymous

  committed
  915fdb5

• remove attention abstraction (comfyanonymous#5324)

  contentis

  authored
  5a8a489

• Optimizations to --fast and scaled fp8.

  comfyanonymous

  committed
  8ce2a10

• Fix duplicate sigmas on beta scheduler.

  comfyanonymous

  committed
  f82314f

Commits on Oct 21, 2024

• Mixed precision diffusion models with scaled fp8.

  Show description for 0075c6d
  comfyanonymous

  committed
  0075c6d

• Support scaled fp8 t5xxl model.

  comfyanonymous

  committed
  83ca891

Commits on Oct 20, 2024

• Fixed model merging issue with scaled fp8.

  comfyanonymous

  committed
  f9f9faf

• fp8 casting is fast on GPUs that support fp8 compute.

  comfyanonymous

  committed
  471cd3e

• Support diffusion models with scaled fp8 weights.

  comfyanonymous

  committed
  a68bbaf

Commits on Oct 19, 2024

• Clamp output when rounding weight to prevent Nan.

  comfyanonymous

  committed
  73e3a9e

Commits on Oct 18, 2024

• Add tooltips to LoraSave node.

  comfyanonymous

  committed
  518c0dc

Commits on Oct 17, 2024

• Add a note that python 3.13 is not yet supported to the README.

  comfyanonymous

  committed
  ce0542e

• Pytorch can be shipped with numpy 2 now.

  comfyanonymous

  committed
  8473019

• Ignore more network related errors during websocket communication. (comfyanonymous#5269)

  Show description for 89f1589
  

  xiaodong-xie

  and

  xxd-frever

  authored
  89f1589

• Use the lowvram cast_to function for everything.

  comfyanonymous

  committed
  6715899

Commits on Oct 16, 2024

• Add missing import.

  comfyanonymous

  committed
  7390ff3

• Revert "Fix Transformers FutureWarning (comfyanonymous#5140)"

  Show description for 0bedfb2
  comfyanonymous

  committed
  0bedfb2

Pagination

PreviousNext