KAT Coder V2.5 Dev OptiQ 4bit is an open-source language model by mlx-community. Features: 34.7b LLM, VRAM: 21.9GB, License: apache-2.0, Quantized, Code Generating, LLM Explorer Score: 0.34.
4-bit 4bit Base model:kwaipilot/kat-coder... Base model:quantized:kwaipilot... Code Codegen Conversational En Mlx Optiq Quantized Qwen3 5 moe Region:us Safetensors Sharded Tensorflow
Model Card on HF 🤗: https://huggingface.co/mlx-community/KAT-Coder-V2.5-Dev-OptiQ-4bit
KAT Coder V2.5 Dev OptiQ 4bit Benchmarks
nn.n% — How the model compares to the reference models: Anthropic Sonnet 3.5 ("so35"), GPT-4o ("gpt4o") or GPT-4 ("gpt4").
What is the LLM Explorer Rank (Score)
KAT Coder V2.5 Dev OptiQ 4bit Parameters and Internals
| LLM Name | KAT Coder V2.5 Dev OptiQ 4bit |
| Repository 🤗 | https://huggingface.co/mlx-community/KAT-Coder-V2.5-Dev-OptiQ-4bit |
| Base Model(s) | |
| Model Size | 34.7b |
| Required VRAM | 21.9 GB |
| Updated | 2026-07-25 |
| Maintainer | mlx-community |
| Model Type | qwen3_5_moe |
| Model Files | |
| Supported Languages | en |
| Quantization Type | 4bit |
| Generates Code | Yes |
| Model Architecture | Qwen3_5MoeForConditionalGeneration |
| License | apache-2.0 |
| Model Max Length | 262144 |
| Transformers Version | 4.57.1 |
| Tokenizer Class | Qwen2Tokenizer |
| Padding Token | <|endoftext|> |
| Errors | replace |
Rank the KAT Coder V2.5 Dev OptiQ 4bit Capabilities
🆘 Have you tried this model? Rate its performance. This feedback would greatly assist ML community in identifying the most suitable model for their needs. Your contribution really does make a difference! 🌟
What open-source LLMs or SLMs are you in search of? 55396 in total.
Original data from HuggingFace, Arena and various public git repos.
Check out Ag3ntum — our secure, self-hosted AI agent for server management.
Release v20260328a