← back to catalog · registered 2026-09-24 00:57

Sol45/VLMGuard-Qwen2.5-VL-7B-JailBreakV-GPT4V

Sol45 7B multimodal
curl -H "Authorization: Bearer $ABL_KEY" \
     "https://abliteration.org/api/v1/models/Sol45%2FVLMGuard-Qwen2.5-VL-7B-JailBreakV-GPT4V"
Response includes
  • classification unknown
  • files 6
  • author_summary 1 models
  • readme_text full
10 credits · hourly refresh · ~4 KB payload Get an API key →
Abliteration classifier · v1.0.0
?
Primary method

Unclassified

No clear signals of an abliteration technique in this model.
Confidence
UNKNOWN
Why this label 1 signal
No classification signals present. This may not be an abliterated model at all - it could be a repackaging, a merge with unrelated goals, or unrelated content that mentions the term.
  • no classification signals present (no abliterated, uncensored, or known producer/method markers)
Refusal direction extraction

No specific extraction method could be identified for this model. The producer either did not document it or used a proprietary pipeline.

What is a refusal direction? →
Downloads · 30-day
0
Likes
0
Model age
today
created 2026-09-24

Metadata

Languages
en
Tags
vlmguard vision-language safety en arxiv:2410.00296 region:us
Total size
3.50 MB
Files
6
Quantizations
1
Registered
2026-09-24 00:57
Last updated on HF
2026-09-24 00:15

Files by quantization

Auxiliary files 6 files 3.54 MB
classifier.pt 3.50 MB cce4c489 download
subspace.npz 30.2 KB e4efe32f download
README.md 2.01 KB 6f22cd13 download
.gitattributes 1.48 KB a6344aac download
metrics.json 745 B d0edab45 download
SHA256SUMS.json 267 B a1c41c05 download

README current version from Hugging Face


language:

  • en
    tags:
  • vlmguard
  • vision-language
  • safety

VLMGuard JailBreakV/GPT4V detector

The VLMGuard detector trained with Qwen2.5-VL-7B-Instruct features for the JailBreakV/GPT4V evaluation. See the paper and VLMGuard code repository. The detector weights do not include the Qwen backbone or datasets.

The checkpoint uses Qwen2.5-VL-7B-Instruct revision cc594898137f460bfe9f0759e9844b3ce807cfb5, final-token head-wise features with shape (N, 28, 3584), selected layer 8 and subspace rank 1. The classifier has hidden width 256 and two output logits.

Metric Value
Held-out AUROC 0.9817013616071427
Held-out AUPR 0.9788129385786769
Held-out examples 9600

classifier.pt contains a PyTorch state dict, subspace.npz contains the selected projection, and metrics.json records the configuration. SHA256SUMS.json records their checksums. These files were converted from the verified training run and tested with the repository's verify_model.py and infer.py; CPU inference on the held-out feature array reproduced AUROC 0.9817013616071427.

Download the four files into one directory, then run:

python verify_model.py --run-dir /path/to/model
python infer.py --run-dir /path/to/model --features /path/to/features.npy --output scores.npy --device cpu

The feature array must use the same backbone revision, token selection, layer indexing and hidden dimension. The output is a maliciousness probability for each input row.

Citation

@article{fang2026vlmguard,
  title={VLMGuard: Bootstrapping Malicious Prompt Detectors from Unlabeled Vision-Language Prompts in the Wild},
  author={Fang, Junlin and Chen, Wenyu and Ghosh, Reshmi and Sim, Robert and Salem, Ahmed and Carvalho, Vitor R. and Lawton, Emily and Li, Sharon and Stokes, Jack W. and Du, Sean},
  journal={Transactions on Machine Learning Research},
  year={2026},
  url={https://openreview.net/forum?id=z7gczmhmmo}
}
Catalog is the map. Apps are the tools.

Run models on your own machine, not in the cloud.

Every model page has an "Open in app" button that hands off directly to a local runtime of your choice - Abliteration, LM Studio, or Ollama. No API keys, no subscription, no prompt leakage.