Runs on-device in the TokForge app.
Pre-converted
Qwen3.5-9B-ultra-uncensored-heretic-v1 in MNN format for on-device inference with
TokForge.
Ultra uncensored heretic variant of Qwen3.5-9B by llmfan46. Fine-tuned for unrestricted, unfiltered conversations with no safety guardrails. Excellent for creative writing, roleplay, and uncensored Q&A.
This model is optimized for
TokForge — a free Android app for private, on-device LLM inference.
Speculative decoding is not recommended for this model. In our testing, draft pairing did not deliver reliable speedups on Qwen3.5 targets.
Actual speed varies by device, thermal state, and generation length. Typical ranges for this model size:
This is an MNN conversion of
Qwen3.5-9B-ultra-uncensored-heretic-v1 by
llmfan46. All credit for the model architecture, training, and fine-tuning goes to the original author(s). This conversion only changes the runtime format for mobile deployment.
