--- language: - en license: apache-2.0 library_name: unsloth tags: - red-team - security - penetration-testing - gemma-4 - qat - mtp - lora - gguf - anthropic-skills base_model: unsloth/Gemma-4-12B-it pipeline_tag: text-generation --- # Gemma 4 12B RedTeam QAT + MTP **This model is superseded by [dazeb2/Gemma-4-12B-redteam-v5](https://huggingface.co/dazeb2/Gemma-4-12B-redteam-v5).** v5 is a significant improvement: 17,271 training records, LoRA r=32, 3 epochs, final loss 0.226. The training dataset was generated from a comprehensive library of **714 genuine Anthropic cybersecurity skills** covering red team operations, penetration testing, and adversary emulation techniques. ## Intended Use This model is designed exclusively for **authorized cybersecurity training and assessment in isolated lab environments** such as [redteamlab.pro](https://redteamlab.pro). Use at your own risk — the user bears full responsibility for compliance with all applicable laws. ## Files | File | Size | Description | |------|------|-------------| | `Gemma-4-12B-redteam-adapter.gguf` | 126 MB | LoRA adapter (GGUF format, legacy) | ## Usage (legacy) Use with the base model from `unsloth/gemma-4-12b-it-qat-GGUF` with `--lora` flag. ## Disclaimer **USE AT YOUR OWN RISK.** The model owner and contributors accept no liability for any damages, losses, or legal consequences arising from the use or misuse of this model. You are solely responsible for complying with all applicable laws and regulations.
