Home Creators Posts Import Register Favorites Logout

Downloads

Content

Video : https://youtu.be/E7lD3oP4TCA

Hey everyone!

New video is up, and this week gave us two open-weight image models within days of each other: Fibo 2.0 from Bria, and MOSS Image from the OpenMOSS team.

I put both into ComfyUI and ran them against the Flux 3 Image Klein from last week. Same prompts, same seeds, nothing dressed up. Three friends doing a coffee shop selfie, a Chinese street sign outside a night market pancake stall, skin and frying oil under real light, a sci-fi greenhouse, and a cyberpunk editorial poster.

The honest split: MOSS Image rendered the Chinese sign correctly while Fibo gave me gibberish, and its JSON layout control is the most interesting thing either model does. But Fibo 2.0 held skin tone and fabric detail through an edit that MOSS visibly smudged, so for pure editing it is not close.

And Flux 3 Klein? Still making faces that look like X-ray scans. I have stopped expecting that to change.

Check out the resources below for the model links, and LORAs.

And attached BBOX and Image Edit Workflows for both Fibo 2.0 and MOSS.



For Patreon Supporters , additional post of these 2 image model post for you guys ( https://www.patreon.com/aifuturetech/posts/173175646 ). Thanks for supporting the channel. You guys make these deep dives possible.

— Benji

Resources:

- Fibo 2.0 Dev (Bria): https://huggingface.co/briaai/Fibo-2.0-dev

- Fibo 2.0 Edit Turbo: https://huggingface.co/briaai/Fibo-2.0-edit-turbo

- Fibo 2.0 ComfyUI repack: https://huggingface.co/Comfy-Org/Fibo-2.0

- MOSS Image Base (fine-tuning): https://huggingface.co/OpenMOSS-Team/MOSS-Image-Base

- MOSS Image SFT (daily generation): https://huggingface.co/OpenMOSS-Team/MOSS-Image-SFT

- MOSS Image ComfyUI repack: https://huggingface.co/Comfy-Org/MOSS-Image

- MOSS Image Edit LoRA: https://huggingface.co/chinoll/Moss-Image-Edit

- ComfyUI support for MOSS Image edit LoRAs: https://github.com/Comfy-Org/ComfyUI/pull/17296

- Workflows from this video (on Patreon): fibo&moss_image_t2i_bboxes.json and fibo&moss_ImageEdit.json

- Previous video — Flux 3 Image Klein raw tests: https://www.youtube.com/watch?v=VtA9vfOVWnU

Files

Fibo 2.0 and MOSS Image in ComfyUI - 2 New Better Open Weight Image Models!

Fibo 2.0 and MOSS Image tested in ComfyUI — two new open-weights models side by side with the Flux 3 Image Klein, same prompts, same seeds, no LoRAs to hide behind. Bria's Fibo 2.0 claims the top open-weights image editor spot and number two in text-to-image. OpenMOSS's MOSS Image is the 5B text-and-layout specialist that got day-one ComfyUI support and then got turned into edit LoRAs by the community within a week. In this video we load both natively, wire the subgraphs, and run all three models through the same tests: a three-character coffee shop selfie, a Chinese street sign outside a night market pancake stall, skin and frying oil under real light, a sci-fi greenhouse, and a cyberpunk editorial poster driven entirely by MOSS Image's JSON bounding-box layout. This is for anyone running ComfyUI locally who wants to know which of these is worth the disk space before downloading 20 GB of weights. You need ComfyUI updated past the day-one support merges and the repacked files from the Comfy-Org repos — every download is linked below. If either model is new to you, the setup section walks the loaders, both text encoders, and the single Flux 2 VAE that all three models happen to share, so you only download the VAE once. Why it matters: this is the first pair of open releases where you can pick by job instead of by benchmark score. MOSS Image renders text and follows a bounding-box layout better than anything in this size class, and it is Apache 2.0 with no strings. Fibo 2.0 keeps skin, fabric and background detail through an edit that the others visibly smudge, and both its Dev and Edit Turbo checkpoints will take a JSON layout prompt even though Bria never advertised that. Neither wins everything, and the same-seed runs show exactly where each one falls apart. Resources: - Fibo 2.0 Dev (Bria): https://huggingface.co/briaai/Fibo-2.0-dev - Fibo 2.0 Edit Turbo: https://huggingface.co/briaai/Fibo-2.0-edit-turbo - Fibo 2.0 ComfyUI repack: https://huggingface.co/Comfy-Org/Fibo-2.0 - MOSS Image Base (fine-tuning): https://huggingface.co/OpenMOSS-Team/MOSS-Image-Base - MOSS Image SFT (daily generation): https://huggingface.co/OpenMOSS-Team/MOSS-Image-SFT - MOSS Image ComfyUI repack: https://huggingface.co/Comfy-Org/MOSS-Image - MOSS Image Edit LoRA: https://huggingface.co/chinoll/Moss-Image-Edit - ComfyUI support for MOSS Image edit LoRAs: https://github.com/Comfy-Org/ComfyUI/pull/17296 - Freebie Workflows from this video : https://www.patreon.com/aifuturetech/posts/173163588?utm_source=youtube&utm_medium=video&utm_campaign=202610 - Previous video — Flux 3 Image Klein raw tests: https://www.youtube.com/watch?v=VtA9vfOVWnU Calliope - Local-first Agentic Idea-to-video Studio https://github.com/benjiyaya/Calliope If You Like tutorial like this, You Can Support Our Work In Patreon: https://www.patreon.com/c/aifuturetech #Fibo2 #MOSSImage #ComfyUI #ImageGeneration #OpenWeights

Comments

No comments found for this post.