What are the hardware requirements for Qwen3.5 397B A17B?

Qwen3.5 397B A17B requires approximately 369.8 GB of VRAM. Quantized variants may run on less VRAM; see the Quantized Models section on this page.

Who developed Qwen3.5 397B A17B and how large is it?

Qwen3.5 397B A17B is developed by Qwen, a model with 397b parameters. The model is published as open weights on Hugging Face and indexed on LLM Explorer with full benchmark history.

How does Qwen3.5 397B A17B perform on standard benchmarks?

Qwen3.5 397B A17B has the following published scores: LMSYS ELO 1442. Compare against reference models on this page or on the LLM Explorer leaderboards.

Qwen3.5 397B A17B by Qwen — VRAM 369.8GB

Name: Qwen3.5 397B A17B
Author: Qwen

Qwen3.5 397B A17B is an open-source language model by Qwen. Features: 397b LLM, VRAM: 369.8GB, License: apache-2.0, LLM Explorer Score: 0.55, ELO: 1442.

Conversational Deploy:azure Endpoints compatible Eval-results Image-text-to-text Qwen3 5 moe Region:us Safetensors Sharded Tensorflow

Model Card on HF 🤗: https://huggingface.co/Qwen/Qwen3.5-397B-A17B

Qwen3.5 397B A17B Benchmarks

LMSys ELO 1442 vs 1272 (so35) 13.4%

^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)

Qwen3.5 397B A17B Parameters and Internals

LLM Name	Qwen3.5 397B A17B
Repository 🤗	https://huggingface.co/Qwen/Qwen3.5-397B-A17B
Model Size	397b
Required VRAM	369.8 GB
Updated	2026-08-09
Maintainer	Qwen
Model Type	qwen3_5_moe
Model Files	8.6 GB: 1-of-94 8.6 GB: 2-of-94 8.6 GB: 3-of-94 8.6 GB: 4-of-94 8.6 GB: 5-of-94 8.6 GB: 6-of-94 8.6 GB: 7-of-94 8.6 GB: 8-of-94 8.6 GB: 9-of-94 8.6 GB: 10-of-94 8.6 GB: 11-of-94 8.6 GB: 12-of-94 8.6 GB: 13-of-94 8.6 GB: 14-of-94 8.6 GB: 15-of-94 8.6 GB: 16-of-94 8.6 GB: 17-of-94 8.6 GB: 18-of-94 8.6 GB: 19-of-94 8.6 GB: 20-of-94 8.6 GB: 21-of-94 8.6 GB: 22-of-94 8.6 GB: 23-of-94 8.6 GB: 24-of-94 8.6 GB: 25-of-94 8.6 GB: 26-of-94 8.6 GB: 27-of-94 8.6 GB: 28-of-94 8.6 GB: 29-of-94 8.6 GB: 30-of-94 8.6 GB: 31-of-94 8.6 GB: 32-of-94 8.6 GB: 33-of-94 8.6 GB: 34-of-94 8.6 GB: 35-of-94 8.6 GB: 36-of-94 8.6 GB: 37-of-94 8.6 GB: 38-of-94 8.6 GB: 39-of-94 8.6 GB: 40-of-94 8.6 GB: 41-of-94 8.6 GB: 42-of-94 8.6 GB: 43-of-94
Model Architecture	Qwen3_5MoeForConditionalGeneration
License	apache-2.0
Transformers Version	4.57.0.dev0

Rank the Qwen3.5 397B A17B 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

Quantized Models of the Qwen3.5 397B A17B

Model	Likes	Downloads	VRAM
Qwen3.5 397B A17B GGUF	253	63434	0 GB
Qwen3.5 397B A17B 8bit	5	105802	209 GB
Qwen3.5 397B A17B GPTQ Int4	35	22114	92 GB

Best Alternatives to Qwen3.5 397B A17B

Best Alternatives	Context / RAM	Downloads	Likes
Qwen3.5 397B A17B NVFP4	0K / 251.1 GB	428565	102
Qwen3.5 397B A17B NVFP4 V2	0K / 243.5 GB	23781	5
Ornith 1.0 397B	0K / 288.5 GB	318396	260
Ornith 1.0 397B FP8	0K / 147.7 GB	597119	183
XYZ Aquila Pro	0K / 225 GB	1805	375
Qwen3.5 397B A17B	0K / 369.8 GB	53	13
Qwen3.5 397B A17B FP8	0K / 184.9 GB	825515	182
Qwen3.5 397B A17B MXFP4	0K / 98.9 GB	21989	6
Ornith 1.0 397B	0K / 288.5 GB	447	4
...7B A17B Heretic Int4 AutoRound	0K / 212.3 GB	4351	4