--- library_name: transformers license: mit pipeline_tag: text-generation base_model: - deepreinforce-ai/Ornith-1.0-9B tags: - gguf - llama.cpp - speculative-decoding - mtp - multi-token-prediction - qwen3.5 - abliterated - uncensored - GGUF - huihui extra_gated_prompt: >- **Usage Warnings** “**Risk of Sensitive or Controversial Outputs**“: This model’s safety filtering has been significantly reduced, potentially generating sensitive, controversial, or inappropriate content. Users should exercise caution and rigorously review generated outputs. “**Not Suitable for All Audiences**:“ Due to limited content filtering, the model’s outputs may be inappropriate for public settings, underage users, or applications requiring high security. “**Legal and Ethical Responsibilities**“: Users must ensure their usage complies with local laws and ethical standards. Generated content may carry legal or ethical risks, and users are solely responsible for any consequences. “**Research and Experimental Use**“: It is recommended to use this model for research, testing, or controlled environments, avoiding direct use in production or public-facing commercial applications. “**Monitoring and Review Recommendations**“: Users are strongly advised to monitor model outputs in real-time and conduct manual reviews when necessary to prevent the dissemination of inappropriate content. “**No Default Safety Guarantees**“: Unlike standard models, this model has not undergone rigorous safety optimization. huihui.ai bears no responsibility for any consequences arising from its use. --- # huihui-ai/Huihui-Ornith-1.0-9B-abliterated-MTP-GGUF This is an uncensored version of [deepreinforce-ai/Ornith-1.0-9B](https://huggingface.co/deepreinforce-ai/Ornith-1.0-9B) created with abliteration (see [remove-refusals-with-transformers](https://github.com/Sumandora/remove-refusals-with-transformers) to know more about it). This is a crude, proof-of-concept implementation to remove refusals from an LLM model without using TransformerLens. **Note** We will directly perform ablation on the GGUF files from [protoLabsAI/Ornith-1.0-9B-MTP-GGUF](https://huggingface.co/protoLabsAI/Ornith-1.0-9B-MTP-GGUF). Some weights quantized in Q3_K, Q4_K, Q5_K and Q6_K will be converted to Q8_0. This time, the weights of the first 5 layers, so the final weights will increase, but the change will not be significant. We performed a simple quantization comparison between Q8_0 and MXFP4 on the Q3 version of protoLabsAI/Ornith-1.0-9B-MTP-GGUF. Clearly, Q8_0 has the largest weight size, but its PPL is smaller than the original. While MXFP4 can reduce the model weight size, it increases the PPL. name | Size | PPL | weight ---------------------------------------------------------------------|--------|--------------------|-------------------------------------- [Ornith-1.0-9B-MTP-GGUF/ornith-9b-mtp-kl-IQ3_M.gguf](https://huggingface.co/protoLabsAI/Ornith-1.0-9B-MTP-GGUF/tree/main) | 4.67GB | 7.2065 +/- 0.04554 | Q6_K(output), IQ3_S(ffn_down/ssm_out) [Huihui-Ornith-1.0-9B-abliterated-MTP-GGUF-Q8/ornith-9b-mtp-kl-IQ3_M.gguf](https://huggingface.co/huihui-ai/Ornith-1.0-9B-abliterated-MTP-GGUF/tree/main) | 5.99GB | **7.1733 +/- 0.04588** | Q8_0(output/ffn_down/ssm_out) Huihui-Ornith-1.0-9B-abliterated-MTP-GGUF/ornith-9b-mtp-kl-IQ3_M.gguf | 4.84GB | 7.2943 +/- 0.04648 | Q6_K(output), Q8_0(ffn_down/ssm_out) Huihui-Ornith-1.0-9B-abliterated-MTP-GGUF-MXFP4/ornith-9b-mtp-kl-IQ3_M.gguf | **4.42GB** | 7.3632 +/- 0.04687 | MXFP4(output/ffn_down/ssm_out) ## Download Use the latest [llama.cpp](https://github.com/ggml-org/llama.cpp), ``` huggingface-cli download huihui-ai/Huihui-Ornith-1.0-9B-abliterated-MTP-GGUF --local-dir ./huihui-ai/Huihui-Ornith-1.0-9B-abliterated-MTP-GGUF --token xxx ``` ### Usage Warnings - **Risk of Sensitive or Controversial Outputs**: This model’s safety filtering has been significantly reduced, potentially generating sensitive, controversial, or inappropriate content. Users should exercise caution and rigorously review generated outputs. - **Not Suitable for All Audiences**: Due to limited content filtering, the model’s outputs may be inappropriate for public settings, underage users, or applications requiring high security. - **Legal and Ethical Responsibilities**: Users must ensure their usage complies with local laws and ethical standards. Generated content may carry legal or ethical risks, and users are solely responsible for any consequences. - **Research and Experimental Use**: It is recommended to use this model for research, testing, or controlled environments, avoiding direct use in production or public-facing commercial applications. - **Monitoring and Review Recommendations**: Users are strongly advised to monitor model outputs in real-time and conduct manual reviews when necessary to prevent the dissemination of inappropriate content. - **No Default Safety Guarantees**: Unlike standard models, this model has not undergone rigorous safety optimization. huihui.ai bears no responsibility for any consequences arising from its use. ### Donation If you like it, please click 'like' and follow us for more updates. You can follow [x.com/support_huihui](https://x.com/support_huihui) to get the latest model information from huihui.ai. ##### Your donation helps us continue our further development and improvement, a cup of coffee can do it. - bitcoin(BTC): ``` bc1qqnkhuchxw0zqjh2ku3lu4hq45hc6gy84uk70ge ``` - Support our work on Ko-fi (https://ko-fi.com/huihuiai)!
