What are the hardware requirements for Qwen3 235B A22B Thinking 2507 FP8?

Qwen3 235B A22B Thinking 2507 FP8 requires approximately 236.5 GB of VRAM and supports a context window of 256K tokens. Quantized variants may run on less VRAM; see the Quantized Models section on this page.

Who developed Qwen3 235B A22B Thinking 2507 FP8 and how large is it?

Qwen3 235B A22B Thinking 2507 FP8 is developed by Qwen, a model with 235b 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 Qwen3 235B A22B Thinking 2507 FP8?

Qwen3 235B A22B Thinking 2507 FP8 is hosted on Hugging Face and linked from this page. LLM Explorer also lists quantized variants and similar alternatives if available.

Qwen3 235B A22B Thinking 2507 FP8 by Qwen — VRAM 236.5GB, 256K context

Name: Qwen3 235B A22B Thinking 2507 FP8
Author: Qwen

Qwen3 235B A22B Thinking 2507 FP8 is an open-source language model by Qwen. Features: 235b LLM, VRAM: 236.5GB, Context: 256K, License: apache-2.0, LLM Explorer Score: 0.26.

Arxiv:2505.09388 Base model:quantized:qwen/qwen... Base model:qwen/qwen3-235b-a22... Conversational Deploy:azure Endpoints compatible Fp8 Qwen3 moe Region:us Safetensors Sharded Tensorflow

Model Card on HF 🤗: https://huggingface.co/Qwen/Qwen3-235B-A22B-Thinking-2507-FP8

Qwen3 235B A22B Thinking 2507 FP8 Parameters and Internals

LLM Name	Qwen3 235B A22B Thinking 2507 FP8
Repository 🤗	https://huggingface.co/Qwen/Qwen3-235B-A22B-Thinking-2507-FP8
Base Model(s)	Qwen3 235B A22B Thinking 2507 Qwen/Qwen3-235B-A22B-Thinking-2507
Model Size	235b
Required VRAM	236.5 GB
Updated	2026-08-01
Maintainer	Qwen
Model Type	qwen3_moe
Model Files	10.0 GB: 1-of-24 10.0 GB: 2-of-24 10.0 GB: 3-of-24 10.0 GB: 4-of-24 10.0 GB: 5-of-24 10.0 GB: 6-of-24 10.0 GB: 7-of-24 10.0 GB: 8-of-24 10.0 GB: 9-of-24 10.0 GB: 10-of-24 10.0 GB: 11-of-24 10.0 GB: 12-of-24 10.0 GB: 13-of-24 10.0 GB: 14-of-24 10.0 GB: 15-of-24 10.0 GB: 16-of-24 10.0 GB: 17-of-24 10.0 GB: 18-of-24 10.0 GB: 19-of-24 10.0 GB: 20-of-24 10.0 GB: 21-of-24 10.0 GB: 22-of-24 10.0 GB: 23-of-24 6.5 GB: 24-of-24
Model Architecture	Qwen3MoeForCausalLM
License	apache-2.0
Context Length	262144
Model Max Length	262144
Transformers Version	4.51.0
Tokenizer Class	Qwen2Tokenizer
Padding Token	<\|endoftext\|>
Vocabulary Size	151936
Torch Data Type	bfloat16
Errors	replace

Rank the Qwen3 235B A22B Thinking 2507 FP8 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 Qwen3 235B A22B Thinking 2507 FP8

Best Alternatives	Context / RAM	Downloads	Likes
...en3 235B A22B Instruct 2507 1M	986K / 171.9 GB	4	1
Qwen3 235B A22B Instruct 2507	256K / 167.9 GB	28	10
Qwen3 235B A22B Thinking 2507	256K / 167.9 GB	14	3
Qwen3 235B A22B Instruct 2507	256K / 171.9 GB	59472	793
Vivo C V1	256K / 156.4 GB	415	2
... 235B A22B Thinking 2507 NVFP4	256K / 139.2 GB	22828	8
... 235B A22B Instruct 2507 NVFP4	256K / 139.2 GB	14322	10
Qwen3 235B A22B Thinking 2507	256K / 171.9 GB	20387	408
...n3 235B A22B Instruct 2507 FP8	256K / 236.5 GB	54126	147
MiroThinker 1.7 Mini	256K / 61.1 GB	584	104

Note: green Score (e.g. "73.2") means that the model is better than Qwen/Qwen3-235B-A22B-Thinking-2507-FP8.