LLM EXPLORER 56,604 MODELS INDEXED

KAT Coder V2.5 Dev Vision OptiQ 4bit by sluttybutfast

By sluttybutfast · 93 downloads

KAT Coder V2.5 Dev Vision OptiQ 4bit is an open-source language model by sluttybutfast. Features: 35b LLM, VRAM: 22.9GB, License: apache-2.0, Quantized, Code Generating.

  4-bit   4bit Base model:mlx-community/kat-c... Base model:quantized:mlx-commu...   Code   Codegen   Conversational   Image-text-to-text   Mlx   Multimodal   Quantized   Qwen3 5 moe   Region:us   Safetensors   Sharded   Tensorflow   Vision   Vision-language

KAT Coder V2.5 Dev Vision OptiQ 4bit Parameters and Internals

LLM NameKAT Coder V2.5 Dev Vision OptiQ 4bit
Repository πŸ€—https://huggingface.co/sluttybutfast/KAT-Coder-V2.5-Dev-Vision-OptiQ-4bit 
Base Model(s)  mlx-community/KAT-Coder-V2.5-Dev-OptiQ-4bit   unsloth/Qwen3.6-35B-A3B-UD-MLX-4bit   mlx-community/KAT-Coder-V2.5-Dev-OptiQ-4bit   unsloth/Qwen3.6-35B-A3B-UD-MLX-4bit
Model Size35b
Required VRAM22.9 GB
Updated2026-08-09
Maintainersluttybutfast
Model Typeqwen3_5_moe
Model Files  4.3 GB: 1-of-6   4.3 GB: 2-of-6   4.3 GB: 3-of-6   4.2 GB: 4-of-6   4.2 GB: 5-of-6   1.6 GB: 6-of-6
Quantization Type4bit
Generates CodeYes
Model ArchitectureQwen3_5MoeForConditionalGeneration
Licenseapache-2.0
Model Max Length262144
Transformers Version4.57.1
Tokenizer ClassQwen2Tokenizer
Padding Token<|endoftext|>
Errorsreplace

Best Alternatives to KAT Coder V2.5 Dev Vision OptiQ 4bit

Best Alternatives
Context / RAM
Downloads
Likes
Qwopus3.6 35B A3B Coder 8bit0K / 37.8 GB7404
Qwopus3.6 35B A3B Coder 4bit0K / 20.5 GB6863
Qwopus3.6 35B A3B Coder 6bit0K / 28.9 GB2562
KAT Coder V2.5 Dev0K / 69.7 GB17885537
Qwopus3.6 35B A3B Coder Bf160K / 69.9 GB2271
KAT Coder V2.5 Dev VL OQ4e Mtp0K / 21.5 GB1491
KAT Coder V2.5 Dev VL OQ6e Mtp0K / 30.2 GB513
KAT Coder V2.5 Dev VL OQ8e Mtp0K / 38.4 GB2982
Note: green Score (e.g. "73.2") means that the model is better than sluttybutfast/KAT-Coder-V2.5-Dev-Vision-OptiQ-4bit.