Edit any photo with a text instruction
Upload a photo, say what to change, and get the result in seconds. You get 3 free images with no signup. Edits run on Bria Fibo-2. Text posters run on MOSS-Image, the open-source model OpenMOSS released on 2026-10-16 under Apache-2.0.
Edits powered by Bria Fibo-2; posters by MOSS-Image
How it works
- 1
Pick a mode and upload a photo
Choose what kind of edit you want. Text posters need no photo.
- 2
Say what to change
Write one clear instruction. Add up to 3 reference images if you need them.
- 3
Generate, compare, download
Drag the slider to compare before and after, then download the result.
Five modes, one example each
Click an example to load it into the editor.
Background swap
Put the product or person on a new background.
“Replace the background with a clean white studio backdrop and soft shadow.”
Color / Outfit
Change colors, clothing or materials.
“Change the jacket to dark green leather; keep the face and pose unchanged.”
Lighting / Season
Day to night, summer to winter, golden hour.
“Make it night with warm street lights and a light snowfall.”
Style
Anime, oil painting, watercolor and more.
“Turn this photo into a clean anime illustration with flat colors.”
Text poster
Generate a poster with readable text (MOSS-Image text-to-image).
“A minimalist coffee shop poster, headline "MORNING ROAST", subline "Fresh daily, 7am–11am", warm beige palette.”
Which model does this use?
Photo edits use Bria Fibo-2 Edit, run through fal.ai. Text posters use MOSS-Image, the OpenMOSS text-to-image model, also through fal.ai. The hosted MOSS-Image model cannot take a photo as input, so edits are not made by MOSS-Image.
Want the real MOSS-Image Edit LoRA? Read how to run it locally in ComfyUI on the MOSS-Image Edit page.
Credit packs
1 credit = 1 output image. After your 3 free images.
Starter
$4.99
50 images
$0.100 per image
Pro
$9.99
120 images
$0.083 per image
Studio
$19.99
300 images
$0.067 per image
No subscription. Credits never expire.
MOSS-Image at a glance
| Developer | OpenMOSS |
|---|---|
| Release | 2026-10-16 |
| Size | 5B DiT |
| License | Apache-2.0 (commercial use OK) |
| Variants | Base, SFT, Kromoss community fine-tune |
| Online here | Posters only |
FAQ
Are the edits made by MOSS-Image?
No. Photo edits on this site run on Bria Fibo-2 Edit via fal.ai. Only text posters run on MOSS-Image, because the hosted MOSS-Image model on fal.ai is text-to-image only and cannot take a photo as input. MOSS-Image Edit LoRAs exist for local ComfyUI; see our MOSS-Image Edit page.
Is MOSS-Image free?
The model is free and open source under Apache-2.0. On this site you get 3 free images with no signup. After that, credit packs start at $4.99.
Can I use the results commercially?
MOSS-Image is Apache-2.0, which allows commercial use of the model. Edits come from Bria Fibo-2 via fal.ai, so check their terms for your use case. You are responsible for having the rights to the photos you upload and to what they show.
How do the 3 free images work?
Each output image counts as one. The free limit is tracked per browser and per network address, so you do not need an account. After that, buy credits or sign in to use credits you already have.
Do credits expire?
No. One credit equals one output image. There is no subscription.
What photo formats and sizes can I upload?
JPG, PNG and WebP, up to 10 MB each. You can add up to 3 optional reference images and refer to them by position in your instruction.
Is MOSS-Image related to moss photos?
No. MOSS-Image is an AI model from OpenMOSS, not pictures of moss plants.