ComfyUI

Timeline

1. Oct 19, 2026 · Open-source release · 2 sources
   Comfy-Org/ComfyUI v0.42.1
   Fix some crashing issues a few users experienced when running the MOSS image template workflow.
2. Oct 19, 2026 · Opinion / analysis · 1 source
   Uncensored prompt upsampler for Moss image
   The comfyui default workflow for Moss image has option to enable prompt upsampler which uses the qwen3.5 9b text encoder.
3. Oct 19, 2026 · Opinion / analysis · 1 source
   More than just talking head videos with Clown Anchoring
   So this video was intended to release clown context but it's actually already been made by Seitanism.
4. Oct 19, 2026 · Opinion / analysis · 1 source
   How good is FLUX.3 Klein at editing? A 12-task comparison that surprised me
   All we've been hearing for the last few days is how FLUX.3 Klein didn't live up to expectations when it comes to text-to-image.
5. Oct 19, 2026 · Opinion / analysis · 1 source
   Struggling with Moss
   I tried it with n$fw prompts I copypasta'd from Kroma 0.3.1 and Krea 2 which were absolutely perfectly done on Krea/Kroma's side but the same prompts I put into Moss are outright shit.
6. Oct 19, 2026 · Opinion / analysis · 1 source
   Basic Workflow Simple Prompt Testing: Krea2 vs MOSS vs Qwen2.1 vs Flux3
   Finally had some time today to tinker around with the latest open source releases, put together a gallery of results.
7. Oct 19, 2026 · Open-source release · 1 source
   sgl-project/sglang v0.5.22
   | Model | Type | Cookbook |
8. Oct 18, 2026 · Opinion / analysis · 1 source
   faster Moss gens for iterating (30s→10s)
   I know Moss comes boxed with a 50 step CFG 3 workflow, but it actually performs pretty well at lower step numbers (e.g. 20 in my tests) with minimal loss in quality and changes in the output.
9. Oct 18, 2026 · Open-source release · 1 source
   Comfy-Org/ComfyUI v0.42.0
10. Oct 18, 2026 · Opinion / analysis · 1 source
    ComfyUI-MOSS-Prompt-Enhancer
    Hey there, I made MOSS Prompt Enhancer, a custom node that brings the MOSS-Image prompt enhancement workflow into ComfyUI.
11. Oct 18, 2026 · Opinion / analysis · 1 source
    More Moss SFT images from the creative vault, testing styles and cameras
    More images of stuff that isn't just generic "person standing / portrait / bleh"
12. Oct 18, 2026 · Opinion / analysis · 1 source
    Generative Fill, Remove and Expand now native in GIMP 3, running on your own ComfyUI and local models
    The thing I missed most after switching from Photoshop to GIMP 3 was Generative Fill and Adobe's AI editing tools.
13. Oct 18, 2026 · Opinion / analysis · 1 source
    Is using an EGPU viable with ComfyUI?
    I am an AI noob and i recently go a rtx 3090 to use with comfyUI.
14. Oct 18, 2026 · Policy / regulation · 1 source
    Krea2: Have the censorship removals disappeared?
    It looks as if the Krea2 uncensoring methods have been censored.
15. Oct 18, 2026 · Opinion / analysis · 1 source
    MOSS SFT model- can do Girl on Grass, Grass on Girl, and Dark photos
    MOSS can actually do dark images, too , it appears.
16. Oct 18, 2026 · Opinion / analysis · 1 source
    LoRa Training
    So i recently got into LoRa training and i am making a character but the trained LoRa does some stupid things.
17. Oct 18, 2026 · Opinion / analysis · 1 source
    Same prompt, 3 models: SDXL (WAI-Illustrious v17) vs Krea 2 turbo vs MOSS-Image-SFT (anime scenes, no LoRA)
    Tested the three side by side using the exact same text prompt (booru tags only, no style prefix, no quality tags, no negative prompt), 4 scenes × 2 seeds, 1024×1024.
18. Oct 18, 2026 · Opinion / analysis · 1 source
    🚀 Just a few of the many improvements in OBVPM Workflow
    🚀 Just a few of the many improvements in OBVPM Workflow 007:
19. Oct 18, 2026 · Opinion / analysis · 1 source
    FastH3 and LoRAs
    But when I connect the model output to Basic Guide and Basic Scheduler, two nodes model outputs (Model Attention Backend and Model Sparse Attention) lose the connection to the Basic Guider and Basic Scheduler which makes my FastH3... well, not fast.
20. Oct 18, 2026 · Opinion / analysis · 1 source
    RAMBO: FIRST BLOOD PART II - Rambo and Co - Animated Scene (AI Generated Scene Recreated with Minimax H3)
    Sequenced the takes together in ArtCraft EffectCraft (yes, the vibe-coded After Effects heheh)
21. Oct 18, 2026 · Opinion / analysis · 1 source
    First time using ComfyUI. Tried Moss
    First time trying out ComfyUI after hearing good things about the MOSS image model on Twitter.
22. Oct 18, 2026 · Opinion / analysis · 1 source
    Most successful MiniMax H3 reference-to-video prompt structure so far (3 img + 1 vid example)
    The official open-weights MiniMax H3 (Ref2VA) model was trained on a rigid six-section schema (subjectdefinitions, summary, retentionanalysis, detaileddescription, overallsoundscape, nondiegeticmusic) paired with exact angle-bracketed asset tags (<Picture 1>, <Picture 2>, <Picture 3>, <Video 1>, <Audio 1>) and internal entity aliases (<Subject 1>, <Subject 2>).
23. Oct 18, 2026 · Opinion / analysis · 1 source
    MOSS-Image comes to Fizgig
    Fizgig trains on MOSS-Image Base, the checkpoint OpenMOSS recommend for LoRAs, and previews on MOSS-Image SFT, the one most people generate with.
24. Oct 17, 2026 · Opinion / analysis · 1 source
    Vertical Banding in MOSS-Image experiments
    Been digging into the vertical banding in MOSS-Image SFT.
25. Oct 17, 2026 · Opinion / analysis · 1 source
    Flux 3 comfyui update latest
    Flux 3 got an update for comfyui UI yesterday After the criticism it's the latest one please check that out first
26. Oct 17, 2026 · Opinion / analysis · 1 source
    Blown away by Minimax H3 sometimes. It nailed this one!!!!
    • The song, made with Suno, lyrics created by Claude.
27. Oct 17, 2026 · Opinion / analysis · 1 source
    ComfyUI-Drama3 is 3-4x faster now
    I've updated the ComfyUI Drama-3 nodes and they're at least 3x faster now!
28. Oct 17, 2026 · Opinion / analysis · 1 source
    MOSS-Image LoRA training is now in AcademiaSD LoRAlab Trainer Studio (Base and SFT, under 6 GB VRAM)
    AcademiaSD LoRAlab Trainer Studio now supports MOSS-Image, the new open text-to-image model (Apache 2.0).
29. Oct 17, 2026 · Opinion / analysis · 1 source
    Portable multi .char support, Upto 4 character in same generation: Minimax H3 & Flux Klein workflows
    I have been working on the consistent character portable format (.char) for a while & couple of reddit users asked me if multi character is possible for image & video generation.
30. Oct 17, 2026 · Opinion / analysis · 1 source
    MOSS Image - samplers and schedulers
    I tested a few samplers and schedulers on the MOSS Image model in addition to Euler.

Often appears with