curl -H "Authorization: Bearer $ABL_KEY" \
"https://abliteration.org/api/v1/models/ogies99%2FGemma-3-Uncensored_v1" 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? →This model's place in the market. Above: what it was derived from. Below: the tree of everything derived from it.
The same weights this author released in different packaging. Pick the format that matches your runtime.
| adapter_model.safetensors | 114 MB | 71d01252 | download |
| tokenizer.json | 31.8 MB | daab2354 | download |
| tokenizer.model | 4.47 MB | 1299c11d | download |
| tokenizer_config.json | 1.10 MB | 63f8d465 | download |
| adapter_config.json | 1.56 KB | 92ebba1c | download |
| .gitattributes | 1.53 KB | 52373fe2 | download |
| chat_template.jinja | 1.50 KB | 7c7339b6 | download |
| README.md | 576 B | 6bd0353b | download |
| processor_config.json | 519 B | 09922022 | download |
base_model: unsloth/gemma-3-4b-it-unsloth-bnb-4bit
tags:
This gemma3 model was trained 2x faster with Unsloth
Every model page has an "Open in Abliteration" button that hands the model directly to the first-party desktop client, at the quantization your rig can actually run. No API keys, no subscription, no prompt leakage.