Qwen3.6-35B-A3B-Uncensored-Genesis-APEX-MTP

Here model: [https://huggingface.co/LuffyTheFox/Qwen3.6-35B-A3B-Uncensored-Genesis-V2-APEX-MTP-GGUF](https://huggingface.co/LuffyTheFox/Qwen3.6-35B-A3B-Uncensored-Genesis-V2-APEX-MTP-GGUF) Safetensors: [https://huggingface.co/LuffyTheFox/Qwen3.6-35B-A3B-Uncensored-Genesis-V2-FP8-Safetensors](https://huggingface.co/LuffyTheFox/Qwen3.6-35B-A3B-Uncensored-Genesis-V2-FP8-Safetensors) *Testing results in Open Code on hardware (Beelink gtr9 pro + Strix Halo) done by my friend on Q8\_K\_P - MTP quant:* 1. 5 sessions with 200k context, not a single glitch, no loops, no repeated tool calls. 2. After 120k tokens he suddenly gave another task that doesn't intersect with what it was doing at all, and it calmly picked up and solved it correctly. 3. Uncensored with MTP support with APEX and APEX Compact quantization. **Recommended quant:** APEX, MTP-APEX **Recommended settings for LM Studio:** [System Prompt](https://huggingface.co/LuffyTheFox/Qwen3.6-35B-A3B-Uncensored-Genesis-V2-APEX-MTP-GGUF/raw/main/System_Prompt.txt) [Chat Template](https://huggingface.co/LuffyTheFox/Qwen3.6-35B-A3B-Uncensored-Genesis-V2-APEX-MTP-GGUF/raw/main/chat_template.jinja) [Chat Template Thinking](https://huggingface.co/LuffyTheFox/Qwen3.6-35B-A3B-Uncensored-Genesis-V2-APEX-MTP-GGUF/raw/main/chat_template_thinking.jinja) Or use this minimal string as the **first line**: >`You are Qwen, created by Alibaba Cloud. You are a helpful assistant.` Then add anything you want after. **Model may underperform without this first line.** Settings: |Parameter|Value| |:-|:-| |Temperature|0.7| |Top K Sampling|20| |Presence Penalty|1.5| |Repeat Penalty|1.0| |Top P Sampling|0.8| |Min P Sampling|0| |Seed|42| Enjoy 😄