library_name: transformers
license: apache-2.0
license_link: https://huggingface.co/Qwen/Qwen3.5-122B-A10B/blob/main/LICENSE
pipeline_tag: image-text-to-text
base_model:
- Qwen/Qwen3.5-122B-A10B
tags: - abliterated
- uncensored
- MTP
- GGUF
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-Qwen3.5-122B-A10B-abliterated
This is an uncensored version of Qwen/Qwen3.5-122B-A10B created with abliteration (see 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.
I needed a better performing model than Qwen3.6-35B to act in Orchestrator, Ask and Architect mode. The model in MTP generates 65 tokens/sec on a multiple RTX 3090 platform.
Probably more if using a lower quant than Q6_K. I have also generated MXFP4_MOE and Q4_K_M quants.
You can export GGML_CUDA_ENABLE_UNIFIED_MEMORY=1 to enable unified memory on linux with Nvidia.
All model quants tested with mmproj.gguf for multi-modal image use.
llama.cpp
Use the latest version of llama.cpp
llama-server --host 0.0.0.0 --jinja --mmap --spec-type draft-mtp --ctx-size 262133 --cache-type-k q4_0 --cache-type-v q4_0 --flash-attn on --model Huihui-Qwen3.5-122B-FULL-Q6_K.gguf --mmproj mmproj.gguf --n-gpu-layers 99 --split-mode layer --repeat-penalty 1.10 --jinja
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 nor I bear any responsibility for any consequences arising from its use.
Donation
Your donation helps us continue our further development and improvement, a cup of coffee can do it and will be much appreciated!
- bitcoin（BTC):
bc1q6nvh39fcmy0de0ezepnn2z0rn4dme9yjal77ah