---
title: gevaertlab/MANTA-27B
url: https://recentmodel.com/models/gevaertlab/MANTA-27B/
status: just landed · new, unvetted
published: 2026-11-10T05:19:41Z
updated: 2026-11-14T00:35:18Z
source: https://huggingface.co/gevaertlab/MANTA-27B
---

# gevaertlab/MANTA-27B

MANTA (Medical Answers, Natively from Text And images) is a medical decision model developed by the Gevaert Lab at the Department of Computational Medicine, Stanford University. Dense, 27B params. Context: 256K tokens (config). License: apache-2.0.


- **Org:** gevaertlab
- **Released:** 2026-11-10
- **Status:** just landed · new, unvetted
- **License:** apache-2.0 (permissive)
- **Parameters:** 27.5B
- **Architecture:** dense · MantaForDecision
- **Context:** 262,144 tokens (config.json)
- **Task:** text-classification · text
- **Signal:** 13 HF likes · 156 downloads

## Runs on

mixed: real files + estimate; sizing calculation, not tested.

- Q4 ~19.6 GB (estimated): 24GB GPU (3090 / 4090), 32GB GPU (5090), 24GB GPU (7900 XTX), Mac 36GB, CPU only, 32GB RAM (slow)
- Q8 ~32.3 GB (estimated): 48GB GPU (RTX 6000 Ada / A6000), Mac 48GB, Mac 64GB (M1 Max), CPU only, 64GB RAM (slow)
- FP16 ~58.1 GB: 96GB GPU (RTX Pro 6000 Blackwell), 1× DGX Spark (128GB), 2× DGX Spark (256GB), Strix Halo 128GB unified (~96GB usable), Mac 96GB, Mac 128GB, Mac 192GB, Mac 256GB, Mac 512GB, CPU only, 128GB RAM (slow), CPU only, 256GB RAM (slow)

## Links

- [Hugging Face](https://huggingface.co/gevaertlab/MANTA-27B)