MultiMatte
An open-source, promptable AI background-removal model — name the object you want to keep and it mattes out everything else.
Last verified · 2026-10-23 · by Moe Ameen
Published 2026-09-10 · Updated
Short answer
MultiMatte is Feyn Labs' open-source background-removal model, published September 10, 2026, that keeps whatever object you name in a text prompt. It fine-tunes Meta's SAM 3 with a rank-16 LoRA that changes 19.49M of about 860M parameters, and it outputs alpha mattes, so hair and fuzzy edges stay soft. Feyn reports a 0.901 S-measure on DIS-VD against SAM 3's 0.667. It is Apache-2.0.
Key takeaways
- MultiMatte removes backgrounds by prompt: name the object to keep and it mattes out the rest.
- It is SAM 3 with a small LoRA fine-tune touching about 2.27% of the weights.
- Alpha mattes give each pixel an opacity value, which keeps hair and fur edges clean.
- Feyn reports 0.901 S-measure on DIS-VD versus SAM 3's 0.667.
- Weights are on Hugging Face as feyninc/multimatte under Apache-2.0, with a browser demo.
What is MultiMatte?
MultiMatte is an open-source model for promptable background removal, released by Feyn Labs and detailed in a technical write-up published on September 10, 2026 by Hafedh Hichri and Shreyash Nigam. It is the follow-up to Feyn's earlier FeyNoBg.
The defining feature is that you aim it with words: instead of keeping whatever it decides is the foreground, MultiMatte lets you name the object you want to keep — the classic example is a photo with a dog and a bowl where you tell it to keep just the dog — and it isolates that subject and mattes out the rest. That makes it genuinely useful on cluttered, real-world shots where "the foreground" is ambiguous.
Under the hood it is built on Meta's SAM 3, the concept-promptable segmentation model from 2025, and inherits SAM 3's text alignment and vocabulary. Feyn's contribution is a small low-rank fine-tune — a rank-16 LoRA adapter modifying roughly 19.49M of SAM 3's ~860M parameters (about 2.27% of the weights) across attention and MLP projections, including the CLIP text tower.
MultiMatte replaces SAM 3's binary segmentation masks with alpha mattes: every pixel gets a continuous opacity value, so fuzzy boundaries like hair, fur, and motion blur are described as they actually appear instead of chopped to a hard edge. Feyn reports the tuning drives a large accuracy jump — for example a 0.901 S-measure on the DIS-VD benchmark against SAM 3's 0.667.
It is released openly. MultiMatte ships through Feyn's NoBg Python library under the Apache-2.0 license, with model weights on Hugging Face under `feyninc/multimatte` and a browser demo at usefeyn.com/multimatte, so you can try one image or self-host it in your own pipeline.
Two honest notes. First, MultiMatte is a still-image model today — the write-up and demo are both built around a single uploaded photo, with no stated video support, so removing a background from a clip means running it frame by frame yourself. Second, it is a model and a library, not a consumer app: the demo takes a single image, and doing anything at volume means code. Treat the specific numbers here as a launch-write-up snapshot and check Feyn's pages for current details.
What can you make with MultiMatte?
- Transparent-PNG cutouts of one named subject pulled out of a busy or crowded scene, with the rest of the frame removed
- Clean alpha mattes where hair, fur, and motion-blurred edges stay intact instead of haloing
- Isolated product shots for e-commerce — the exact item selected by prompt, ready to drop onto any background
- Precise subject cutouts for thumbnails, ad layouts, and composites where you need one element and not the whole foreground
- A self-hosted, promptable background-removal step wired into your own image pipeline via the open NoBg library
What makes MultiMatte better than an ordinary background remover?
MultiMatte's edge over an ordinary remover is precision by prompt: it will pull one named subject out of a messy frame — the founder out of a busy office, one SKU off a crowded shelf — with hair-clean edges. That precision is exactly what makes it a great feeder for the image formats where the subject has to be perfectly isolated before anything else works.
Take a MultiMatte cutout into Kompozy as the subject of a Persona Photo, a Photo Post, a Quote Graphic, or a brand-exact Carousel: HyperFrames composites your isolated subject onto your brand background, layers on-style captions and offer or price text, and reframes each piece to 9:16, 1:1, and 16:9 so a single clean cutout becomes native art for every platform.
Where this pays off most is real, imperfect source photos. UGC, phone shots, and event photos rarely have a tidy single-subject foreground — MultiMatte's promptable keep lets you specify the one thing that matters, and Kompozy turns that into a full unit of content: the subject arranged across a Carousel of feature slides, a Quote Graphic pulling a testimonial over it, plus Text Posts and short caption threads written in your voice through a Persona Brief.
Then Autopilot and a per-post review pipeline schedule and publish the set across the eight social platforms plus blog and email from one queue. MultiMatte nails the hard cutout; Kompozy turns it into a week of branded, captioned, scheduled posts.
- Run your photo through MultiMatte (its demo at usefeyn.com/multimatte or the NoBg library) and prompt it to keep the exact subject you want.
- Bring the transparent-PNG cutout into Kompozy and pick formats — Persona Photo, Photo Post, Quote Graphic, or Carousel.
- Let HyperFrames composite the isolated subject onto your brand background with on-style captions and offer text.
- Set a Persona Brief so the captions, text posts, and threads around the image hold your voice.
- Reframe each piece to 9:16, 1:1, and 16:9 and schedule the set across eight social platforms plus blog and email with Autopilot.
What does this look like in Kompozy?
A real photo post made in the Kompozy workspace. It shows the format a clean subject cutout can be placed into; it was not made with MultiMatte.
Who is this for, and who should skip it?
Best for:
- Pulling one named subject out of a busy photo, like a single product on a cluttered table
- Developers who want a self-hosted, promptable background-removal step
- Cutouts where hair, fur or motion blur need soft edges
Not ideal for:
- Video, since the write-up and demo only cover single images
- Non-technical users who need batch processing without writing code
Frequently asked questions
What is MultiMatte?
MultiMatte is an open-source, promptable background-removal model from Feyn, detailed in a September 10, 2026 write-up. You name the object you want to keep and it isolates that subject and mattes out everything else. It is built on Meta's SAM 3 with a small low-rank fine-tune and outputs alpha mattes — a continuous opacity value per pixel — so fuzzy edges like hair cut cleanly.
How is MultiMatte different from a normal background remover?
A normal remover keeps whatever it judges to be the foreground. MultiMatte is promptable: you specify which object to keep by naming it, so it can isolate a single subject out of a cluttered scene rather than retaining every foreground element. That control is its main advantage on busy, real-world photos.
Is MultiMatte free and open source?
Yes. MultiMatte ships through Feyn's NoBg library under the Apache-2.0 license, with weights on Hugging Face (feyninc/multimatte) and a browser demo at usefeyn.com/multimatte. Your only cost is your own compute and integration time. Check Feyn's pages for the exact terms before commercial use.
Does MultiMatte work on video?
Not directly — the write-up and demo are both built around a single uploaded photo, with no stated video support. Removing a background from a clip today means running the model frame by frame through a pipeline you build.
How do I turn a MultiMatte cutout into social posts?
MultiMatte gives you a clean, prompt-isolated cutout but does not publish it. Bring the transparent PNG into Kompozy to build a Persona Photo, Photo Post, Quote Graphic, or brand-exact Carousel on top of it — HyperFrames composites it onto your brand background with captions — then reframe per platform and schedule across eight social platforms plus blog and email from one queue.
How do we know this?
We checked every figure, date, price and quote on this page against the sources listed below on 2026-10-23, and removed or corrected anything we could not confirm. Model details come from Feyn's write-up and the Hugging Face model card. Kompozy plan and format details come from kompozy.io/pricing.
Sources
- Feyn — MultiMatte: background removal with text prompts (read 2026-10-23)
- Hugging Face — feyninc/multimatte (read 2026-10-23)
- Kompozy — Pricing (read 2026-10-23)
Related tools
- FeyNoBg — An open-source, state-of-the-art AI model that automatically removes the background from an image and cuts the subject out onto transparency.
- Refloow Photo Studio — A free, open-source offline photo editor with on-device background removal.
- Moebius — A 0.2B-parameter image inpainting model that claims to match 10B-scale quality at a fraction of the size.
- Adobe AI Studio — A suite of AI editing tools inside Adobe Stock that lets you reshape an asset before you license it.
- Nano Banana 2 Lite — Google's fastest, cheapest Nano Banana image model — a 4-second generator built for high-volume creation.