MANTA decision model
MANTA reads medical text and images in two open model sizes. Check its research use guidance, package limits and maker-reported evaluations.
By Gevaert Lab · Hugging Face weight release: 10 Nov 2026
Medical decision models in two sizes, built on Qwen3.5-4B and Qwen3.8-27B with a trained decision head. They read text and images and return probabilities for each supplied option. These are research models; the card requires independent validation and clinician review before patient decisions.
Quick answers
What is MANTA by Gevaert Lab?
MANTA: Medical decision models in two sizes, built on Qwen3.5-4B and Qwen3.8-27B with a trained decision head.
Who makes MANTA?
MANTA: Gevaert Lab
Facts
- Sizes
- 4B and 27B backbones; the 4B package stores 5,287,287,300 parameters, including a copy of the tied output embedding
- Inputs
- Text or JSON with optional images; up to 64 questions and 4,096 prompt tokens
- Access
- BF16 weights, custom Transformers code and a Python request wrapper
- Licence
- Apache-2.0 files included; source datasets retain their own terms
- Memory
- The maker reports about 16 GB GPU memory for 4B and 60 GB for 27B
Claims and sources
- The maker reports 86.1% accuracy for both sizes on the same 6,700-question medical comparison subset. On its 554 out-of-domain questions, 4B scores 71.5% and 27B 74.4%. These are constructed benchmark questions, not prospective patient outcomes. vendor reportedsource
- The maker reports median request latency of 68 ms for 4B and 148 ms for 27B on one H100 at batch size 1. Hosted-model timings in the card exclude client networking, so they are not end-to-end comparisons. vendor reportedsource
Links
Source: 4B card and weights → Source: 27B card and weights → Source: 4B configuration → Source: Licence →
Related projects
No related projects are listed in the registry.