mentat-sys1-v0.1 is an open-weight model from Yun Qu. Its published files total 490.3 MB.
c72687243aaa8c38f984091247619c242f3f3443a1f3d35c9c3526c4dfde2d83 c2c5b2130fcb6604ab7bb7048464fb7c01281a929df767525525870b67ea7d7d 01ac7306e8312beb749090ad5e3887dc23606d7a5d7dca5d2aa85ff99b612b33 This is an independently trained model.
Model Card
c72687243aaa8c38f984091247619c242f3f3443a1f3d35c9c3526c4dfde2d83 c2c5b2130fcb6604ab7bb7048464fb7c01281a929df767525525870b67ea7d7d 01ac7306e8312beb749090ad5e3887dc23606d7a5d7dca5d2aa85ff99b612b33 This is an independently trained model. It was not initialized from or resumed from another decision-model adapter. The first command downloads the frozen model-specific package into a local manifests, checksums, and aggregate audit evidence. It does not download the Qwen base weights. The second command downloads the separately licensed, pinned base revision needed to serve the model. Installation and serving commands are in the The model answers bounded noul, choice, and score decisions through a…
Excerpt from the card by Yun Qu.
Identity and Version
- Repository
- yunqu/mentat-sys1-v0.1
- Publisher
- Yun Qu
- Task
- Not stated by the source
- Modality
- Other
- Library
- Not stated by the source
- Parameters
- Not stated by the source
- Languages
- Not stated by the source
- Revision
- f6d7f3679c4eaeb3bbaa208b5ed81db05e108bf0
- First published
- 2026-10-04
- Last updated
- 2026-10-04
Files and Weights
20 files, 490.3 MB in total. The weights are 2 files totalling 490.3 MB in safetensors.
Every file
| File | Type | Size | SHA-256 |
|---|---|---|---|
| model/adapter_model.safetensors | Weights | 487.6 MB | c72687243aaa |
| model/decision_readout.safetensors | Weights | 2.6 MB | c2c5b2130fcb |
| evidence/calibrated-model-receipt.json | Configuration | 1.1 KB | — |
| evidence/calibration-report.json | Configuration | 855 B | — |
| evidence/contamination-report.json | Configuration | 5.3 KB | — |
| evidence/legacy-public-evaluation.json | Configuration | 1.4 KB | — |
| model/adapter_config.json | Configuration | 1.3 KB | — |
| model/calibration.json | Configuration | 855 B | — |
| model/decision_readout.json | Configuration | 14.4 KB | — |
| model/model-manifest.json | Configuration | 1.5 KB | — |
| release-manifest.json | Configuration | 4.5 KB | — |
| JEVBENCH_SUBMISSION.md | Documentation | 4.7 KB | — |
| LICENSE | Documentation | 10.8 KB | — |
| MODEL_CARD.md | Documentation | 3.7 KB | — |
| NOTICE | Documentation | 899 B | — |
| README.md | Documentation | 7.4 KB | — |
| REPRODUCIBILITY.md | Documentation | 2.6 KB | — |
| SHA256SUMS | Other | 1.6 KB | — |
| model/SHA256SUMS | Other | 530 B | — |
| .gitattributes | Repository | 1.5 KB | — |
License and Download
- License
- Not stated by the source
- Access
- Open weights, no gate
- Download size
- 490.3 MB
Released by Yun Qu through its official repository on Hugging Face.
Memory Requirements
| Precision | Weights in memory |
|---|---|
| As published | 490.3 MB |
Weights only, from the published parameter count; the key-value cache and runtime add to this.