What are the hardware requirements for KAT Coder V2.5 Dev VL OQ8e Mtp?

KAT Coder V2.5 Dev VL OQ8e Mtp requires approximately 38.4 GB of VRAM. Quantized variants may run on less VRAM; see the Quantized Models section on this page.

Who developed KAT Coder V2.5 Dev VL OQ8e Mtp and how large is it?

KAT Coder V2.5 Dev VL OQ8e Mtp is developed by jason-schulz, a model with 35b parameters. The model is published as open weights on Hugging Face and indexed on LLM Explorer with full benchmark history.

Where can I download or evaluate KAT Coder V2.5 Dev VL OQ8e Mtp?

KAT Coder V2.5 Dev VL OQ8e Mtp is hosted on Hugging Face and linked from this page. LLM Explorer also lists quantized variants and similar alternatives if available.

KAT Coder V2.5 Dev VL OQ8e Mtp by jason-schulz — VRAM 38.4GB

Name: KAT Coder V2.5 Dev VL OQ8e Mtp
Author: jason-schulz

KAT Coder V2.5 Dev VL OQ8e Mtp is an open-source language model by jason-schulz. Features: 35b LLM, VRAM: 38.4GB, License: apache-2.0, Code Generating.

Arxiv:2607.05471 8-bit Agent Agentic-coding Apple-silicon Base model:kwaipilot/kat-coder... Base model:quantized:kwaipilot... Code Codegen Coding Conversational En Image-text-to-text Mlx Moe Mtp Omlx Qwen3 5 moe Region:us Safetensors Sharded Speculative-decoding Tensorflow Vision Zh

Model Card on HF 🤗: https://huggingface.co/jason-schulz/KAT-Coder-V2.5-Dev-VL-oQ8e-mtp

KAT Coder V2.5 Dev VL OQ8e Mtp Parameters and Internals

LLM Name	KAT Coder V2.5 Dev VL OQ8e Mtp
Repository 🤗	https://huggingface.co/jason-schulz/KAT-Coder-V2.5-Dev-VL-oQ8e-mtp
Base Model(s)	Kwaipilot/KAT-Coder-V2.5-Dev Qwen/Qwen3.6-35B-A3B Kwaipilot/KAT-Coder-V2.5-Dev Qwen/Qwen3.6-35B-A3B
Model Size	35b
Required VRAM	38.4 GB
Updated	2026-08-09
Maintainer	jason-schulz
Model Type	qwen3_5_moe
Model Files	5.2 GB: 1-of-8 5.1 GB: 2-of-8 5.1 GB: 3-of-8 5.1 GB: 4-of-8 5.1 GB: 5-of-8 5.1 GB: 6-of-8 5.1 GB: 7-of-8 2.6 GB: 8-of-8
Supported Languages	en zh
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 VL OQ8e Mtp capabilities

Have you tried this model? Rate its performance — this feedback helps the ML community find the right model for their needs.

Instruction Following and Task Automation
Factuality and Completeness of Knowledge
Censorship and Alignment
Data Analysis and Insight Generation
Text Generation
Text Summarization and Feature Extraction
Code Generation
Multi-Language Support and Translation

Best Alternatives to KAT Coder V2.5 Dev VL OQ8e Mtp

Best Alternatives	Context / RAM	Downloads	Likes
KAT Coder V2.5 Dev	0K / 69.7 GB	17885	537
Qwopus3.6 35B A3B Coder Bf16	0K / 69.9 GB	227	1
KAT Coder V2.5 Dev VL OQ4e Mtp	0K / 21.5 GB	149	1
KAT Coder V2.5 Dev VL OQ6e Mtp	0K / 30.2 GB	51	3
Qwopus3.6 35B A3B Coder 8bit	0K / 37.8 GB	740	4
Qwopus3.6 35B A3B Coder 4bit	0K / 20.5 GB	686	3
Qwopus3.6 35B A3B Coder 6bit	0K / 28.9 GB	256	2
...der V2.5 Dev Vision OptiQ 4bit	0K / 22.9 GB	93	2

Note: green Score (e.g. "73.2") means that the model is better than jason-schulz/KAT-Coder-V2.5-Dev-VL-oQ8e-mtp.