Open-source 2B decision model from AWS's Strands Agents project that picks options and returns scores with a reliability figure
Ming-Image-0.1-Design-Layer
New· 4Open-weight 6B model from inclusionAI that splits a flattened design into editable transparent layers, under the MIT licence
About Ming-Image-0.1-Design-Layer
Ming-Image-0.1-Design-Layer approaches design files from the opposite end to an image generator. Instead of producing a picture, it takes a finished, flattened design and recovers the pieces it was built from, handing back a background and each foreground element as its own transparent image. It is an open-weight model of 6 billion parameters from inclusionAI, the research lab of Ant Group, and it is released under the MIT licence alongside a companion model, Ming-Image-0.1-Design. Reports put its release at 22 September 2026.
From one flat image to separate RGBA layers
The input is a single flattened design plus a written plan naming the layers wanted. The output is one transparent RGBA image per layer, each at the same dimensions as the original. The practical purpose is editing: generated or older artwork can be opened in layer-based design tools and adjusted piece by piece, without anyone rebuilding the layers by hand.
Hardware and settings
inclusionAI has validated the model on one GPU that has 80 GB of memory, using BF16 precision. Output is produced at 1024 pixels on a side, and there is a quicker setting at 512.
What is measured and what is not
inclusionAI publishes results on the Crello design test set, measuring how accurately it reproduces colour and transparency. The lab says the model comes out ahead in every setting it tested. These are the developer's own measurements. No independent benchmark was available at release, and the developer offers no hosted API, so anyone wanting to try it runs the weights themselves.
TrustList summary
Ming-Image-0.1-Design-Layer is an open-weight model from inclusionAI, the Ant Group research lab, that works in the reverse direction to an image generator. Given one flattened design and a text plan of the layers wanted, it returns each layer as a separate transparent image. It has 6 billion parameters and was released under the MIT licence alongside Ming-Image-0.1-Design.
What Ming-Image-0.1-Design-Layer does
- Layer splitting: separates a finished design into a background and individual foreground elements, as the plan asks.
- Transparent output: each layer comes back as a separate transparent RGBA image at the input's dimensions.
- Editing legacy artwork: makes generated or older artwork editable in layer-based design tools, without rebuilding it by hand.
- Resolution options: works at 1024 pixels, with a faster 512-pixel option.
Key facts
- Licence: MIT.
- Developer: inclusionAI, the research lab of Ant Group.
- Validated setup: a single GPU with 80 GB of memory in BF16.
- Evaluation: the Crello design test set, where the developer says the model leads every setting it evaluated. No independent benchmark was available at release.
- Hosted access: no hosted API from the developer.
Summary written by TrustList from the business's own published information, Nov 2026.
Benchmarks & AI stats
| Benchmark | Official | Community avg |
|---|---|---|
| No benchmark scores yet. Be the first to add one. | ||
“Official” values are editor-approved and feed the ranking. “Community avg” is the mean of member submissions (shown for transparency; it never affects the ranking until an editor approves a value).
Sign in to add a benchmark score for this model.
Ming-Image-0.1-Design-Layer launches
Trust Score
An earned signal from verification, reviews, awards, transparency and engagement - the vendor can't buy it.
Updated 9/24/2026
Request a demo or quote from Ming-Image-0.1-Design-Layer
Reviews
Write the first review of Ming-Image-0.1-Design-Layer
Used it? Your experience helps other buyers decide.
Write a reviewQuestions & answers
No questions yet. Be the first to ask about Ming-Image-0.1-Design-Layer.
Ming-Image-0.1-Design-Layer alternatives
Similar listings buyers compare against this one.
Google's open multimodal embedding model that maps text, images, audio and video into one vector space, under Apache 2.0
Liquid AI's open-weight decision model, giving yes or no, choice or score answers from text and images
Cantina Security's first open-weight security model, a post-trained GLM-5.3-Flash for vulnerability research and exploit checks
Open-weight translation model from Bilibili's Index team, covering 150 languages and following glossary and style instructions
Perplexity's open-weight decision model that picks among given options for routing and checks in AI systems
Microsoft AI's text-to-speech model for keeping one voice across 23 languages, with voice cloning from short audio
Black Forest Labs' image generation and editing model with bounding-box layouts, multiple reference images and 4K output
More on TrustList
Everything here links back to the same verified catalogue. Pick your next stop.
- Write a reviewHelp other buyers decide
- CompaniesAgencies, consultancies and IT service providers, ranked by verified reviews.
- ProductsSoftware and SaaS with pricing, features, integrations and alternatives.
- ArticlesGuides, comparisons and research from the editorial desk and the community.
- AwardsAnnual recognition decided by verified reviews and an independent jury.
- LaunchesNew products and releases, voted up by the community every day.
- RequestsBuyers describe what they need; vendors respond directly.
- PeopleReviewers, authors and makers with public profiles.
- ComparePut up to four listings side by side before you shortlist.