| base_model: Kwaipilot/KAT-Coder-V2.5-Dev | |
| [EXL3](https://github.com/turboderp-org/exllamav3) quantization of [KAT-Coder-V2.5-Dev](https://huggingface.co/Kwaipilot/KAT-Coder-V2.5-Dev) with the MTP head from [Qwen3.6-35B-A3B](https://huggingface.co/Qwen/Qwen3.6-35B-A3B), 4 bits per weight. | |
| Quant not recommended, try [the non-MTP version](https://huggingface.co/darrowoflykos/Kwaipilot_KAT-Coder-V2.5-Dev-EXL3-4bpw) instead. | |