--- title: FeyNoBg: A SOTA Model For Background Removal | daily.dev description: Feyn introduces FeyNoBg, a state-of-the-art background removal model that leads on four of eight benchmarks and stays within 2% of the leader on the rest.... canonical: https://daily.dev/posts/feynobg-a-sota-model-for-background-removal-30qpjo59b twitter:card: summary_large_image twitter:site: @dailydotdev og:type: website og:site_name: daily.dev og:title: FeyNoBg: A SOTA Model For Background Removal | daily.dev og:description: Feyn introduces FeyNoBg, a state-of-the-art background removal model that leads on four of eight benchmarks and stays within 2% of the leader on the rest.... og:url: https://daily.dev/posts/feynobg-a-sota-model-for-background-removal-30qpjo59b og:image: https://api.daily.dev/og/posts/30qpJo59B.png og:image:alt: FeyNoBg: A SOTA Model For Background Removal og:image:width: 1200 og:image:height: 630 og:locale: en --- > ## Documentation Index > Fetch the complete documentation index at: https://daily.dev/llms.txt > Use this file to discover all available pages before exploring further. # FeyNoBg: A SOTA Model For Background Removal **[Hacker News](https://daily.dev/sources/hn)** · 7 min read · 0 upvotes · 0 comments ## Summary Feyn introduces FeyNoBg, a state-of-the-art background removal model that leads on four of eight benchmarks and stays within 2% of the leader on the rest. Built on BiRefNet, the model was expanded from 222M to 263M parameters by deepening the third feature extractor stage, preserving pretrained weights while adding capacity. Training used a curated mix of 26K+ images from 10 datasets covering camouflage, portraits, high-resolution subjects, and more, with careful balancing to avoid specialization. Alongside the model, Feyn releases NoBg, an open-source Python library providing a unified interface to run and train background removal models, with better throughput and lower GPU memory than the original BiRefNet implementation. Both FeyNoBg and NoBg are available on Hugging Face and GitHub. ## Full article daily.dev links to this article rather than hosting it. Read it at the original source: ## Community take How the wider developer community reacted, aggregated from 1 discussion and 36 comments across hackernews (as of 2026-07-27). **TL;DR:** The community is generally impressed with FeyNoBg's background removal quality and ease of use, but there is notable concern about the cc-by-nc-4.0 license applied to weights derived from an MIT-licensed model (BiRefNet), which limits adoption in professional/commercial contexts. **Sentiment:** 55% positive · 30% mixed · 15% skeptical **The case for** - Background removal is seen as a core, high-value task and progress in this space is welcomed. - The tool is praised for being simple to use compared to alternatives like Segment Anything. - The model handles high-resolution images (beyond 1024x1024) well by resizing the opacity mask. - Community interest in porting the model to mobile devices signals enthusiasm for the project. **The pushback** - Applying cc-by-nc-4.0 to weights derived from an MIT-licensed model raises legal questions and discourages professional/commercial adoption. - The model is fully automatic with no prompt-based subject selection yet, limiting use cases where only a specific foreground element is desired. - A commenter's post was flagged for suspected AI-generated content, raising minor credibility concerns about the submission. **By community** - hackernews (mixed): Commenters are enthusiastic about the technology and its potential but divided over the restrictive cc-by-nc license choice on top of an MIT base model. **Hottest debate:** Whether it is appropriate to release weights fine-tuned from an MIT-licensed model (BiRefNet) under the more restrictive cc-by-nc-4.0 license. **Open questions** - How was the training dataset assembled, particularly given the different training mixes mentioned in the controlled eval? - How does FeyNoBg compare quantitatively to Adobe's background removal model? - When will prompt-based (non-automatic) subject selection be available? **Highlights** > Why extend an MIT licensed model's weights (BiRefNet) and release it under a cc-by-nc-4.0 license? > — [woadwarrior01 on hackernews · 3 comments](https://news.ycombinator.com/item?id=49074829) > This pertains to the larger open source licensing discussions that have been happening (as I'm sure you've seen too). We've released projects under the MIT license before, most notably https://github.com/feyninc/chonkie. While we're not trying to directly monetize on this work, credit goes a long way and helps in other operations. Recently, attributed usage is shrinking. To be clear, this is not a shrinkage in actual use of our software, just how many people acknowledge that they rely on it. cc-by-nc is a protection against that. We've been very honest about our work being on top of BiRefNet as we want to extend the original creators the same courtesy. I have no issues if individuals fork/finetune/or otherwise build on top of any open source projects we release, irrespective of license. At minimum, we want acknowledgment if a company chooses to use our software in production. > — [snyy on hackernews](https://news.ycombinator.com/item?id=49075234) > Your question touches on excellent points. > what is the subject? FeyNoBg is an "automatic" model. It automatically detects foreground elements and segments the image. Most of the time, this includes all foreground elements. As you can see in the freekick example (https://drive.google.com/file/d/1MZkAGLwbhNVOZ0Oi7XvpCfSEu9Q...), the model output includes the ball, Messi, and the Liverpool defenders. In your example, FeyNoBg will segment around the person plus the couch. > I only want the person [including props] The alternative to automatic models are prompt models and those serve the exact use case you're describing. These allow you to specify the foreground element to include. Everything else is removed. That's the next step for FeyNoBg, converting it from an automatic to a prompt model. Now, answering your question: > How does it stack up against Adobe's model? We're better on automatic background removal. Support for selecting a subset of foreground elements is coming soon > — [snyy on hackernews · 1 comments](https://news.ycombinator.com/item?id=49075834) > Btw if you do need a permissively licensed model for this task, there are lots out there, e.g.: https://github.com/KupynOrest/s3od https://github.com/Tennine2077/PDFNet https://github.com/PramaLLC/BEN > — [daemonologist on hackernews](https://news.ycombinator.com/item?id=49075674) **Source threads** - [hackernews](https://news.ycombinator.com/item?id=49072462) · 58 points · 36 comments ## Similar posts on daily.dev - [Seamless Image Background Removal with withoutbg](https://daily.dev/posts/seamless-image-background-removal-with-withoutbg-wiwc70sbz) · Aggregata · 0 upvotes · 0 comments --- Tags: [#machine-learning](https://daily.dev/tags/machine-learning), [#python](https://daily.dev/tags/python), [#computer-vision](https://daily.dev/tags/computer-vision) [View this post on daily.dev](https://daily.dev/posts/feynobg-a-sota-model-for-background-removal-30qpjo59b) ```json {"@context":"https://schema.org","@graph":[{"@type":"Organization","@id":"https://daily.dev/#organization","name":"daily.dev","url":"https://daily.dev","logo":{"@type":"ImageObject","url":"https://daily.dev/apple-touch-icon.png","width":180,"height":180},"sameAs":["https://twitter.com/dailydotdev","https://github.com/dailydotdev","https://www.linkedin.com/company/daily-dev-ltd"]},{"@type":"WebSite","@id":"https://daily.dev/#website","url":"https://daily.dev","name":"daily.dev","publisher":{"@id":"https://daily.dev/#organization"},"potentialAction":{"@type":"SearchAction","target":{"@type":"EntryPoint","urlTemplate":"https://daily.dev/search?q={search_term_string}"},"query-input":"required name=search_term_string"}}]} {"@context":"https://schema.org","@type":"TechArticle","headline":"FeyNoBg: A SOTA Model For Background Removal","url":"https://daily.dev/posts/feynobg-a-sota-model-for-background-removal-30qpjo59b","mainEntityOfPage":{"@type":"WebPage","@id":"https://daily.dev/posts/feynobg-a-sota-model-for-background-removal-30qpjo59b"},"datePublished":"2026-07-27T18:40:54.790Z","dateModified":"2026-07-27T22:17:58.621Z","description":"Feyn introduces FeyNoBg, a state-of-the-art background removal model that leads on four of eight benchmarks and stays within 2% of the leader on the rest....","image":"https://media.daily.dev/image/upload/f_auto,q_auto/v1/posts/68c02649547ba2f6c3fc071d581326c7?_a=AQAEuop","thumbnailUrl":"https://media.daily.dev/image/upload/f_auto,q_auto/v1/posts/68c02649547ba2f6c3fc071d581326c7?_a=AQAEuop","isAccessibleForFree":true,"articleSection":"Hacker News","inLanguage":"en","publisher":{"@type":"Organization","name":"daily.dev","url":"https://daily.dev","logo":{"@type":"ImageObject","url":"https://daily.dev/apple-touch-icon.png","width":180,"height":180}},"author":{"@type":"Organization","name":"Hacker News","logo":"https://media.daily.dev/image/upload/t_logo,f_auto/v1/logos/hn","url":"https://daily.dev/sources/hn"},"commentCount":0,"discussionUrl":"https://daily.dev/posts/feynobg-a-sota-model-for-background-removal-30qpjo59b","interactionStatistic":[{"@type":"InteractionCounter","interactionType":{"@type":"LikeAction"},"userInteractionCount":0},{"@type":"InteractionCounter","interactionType":{"@type":"CommentAction"},"userInteractionCount":0}],"keywords":"machine-learning,python,computer-vision","timeRequired":"PT7M"} {"@context":"https://schema.org","@type":"BreadcrumbList","itemListElement":[{"@type":"ListItem","position":1,"name":"Home","item":"https://daily.dev"},{"@type":"ListItem","position":2,"name":"Hacker News","item":"https://daily.dev/sources/hn"},{"@type":"ListItem","position":3,"name":"FeyNoBg: A SOTA Model For Background Removal"}]} ```