Qwen3-235B-A22B

Name: Qwen3-235B-A22B
Price: 0.002 CNY
Availability: InStock
Author: Alibaba

Reasoning modelQwen3

Qwen3-235B-A22B

Release date: 2025-04-28Updated: 2026-07-17 23:18:26.2562,744

Live demo GitHub Hugging Face Compare

Parameters

235B

Context length

128K

Chinese support

Supported

Reasoning ability

Qwen3-235B-A22B is a reasoning model from Alibaba, released on 2025-04-28. It accepts text input and returns text output. The cataloged parameter count is 235B, with 22B active parameters per inference. The recorded context window is 128K, and the recorded maximum output is 16K. Cataloged capabilities include Reasoning model and Multilingual. The checkpoint is listed under the Apache 2.0 license, with the license link included in the model references. Use the linked references to confirm current access, licensing, and provider-specific limits.

Data sourced primarily from official releases (GitHub, Hugging Face, papers), then benchmark leaderboards, then third-party evaluators. Learn about our data methodology

Qwen3-235B-A22B

Model basics

Reasoning traces

Supported

Thinking modes

Thinking modes not supported

Context length

128K tokens

Max output length

16K tokens

Model type

Reasoning model

Modality (in / out)

Text → Text

Release date

2025-04-28

Model file size

470GB

MoE architecture

Yes

Total params / Active params

235B / 22B

Knowledge cutoff

No data

Qwen3-235B-A22B

Open source & experience

Code license

Apache 2.0

Weights license

Apache 2.0- Commercial use permitted

GitHub repo

https://github.com/QwenLM/Qwen3

Hugging Face

https://huggingface.co/Qwen/Qwen3-235B-A22B

Live demo

https://chat.qwen.ai/

Qwen3-235B-A22B

Official resources

Paper

DataLearnerAI blog

Qwen3-235B-A22B

API details

API speed

3/5

💡Default unit: $/1M tokens. If vendors use other units, follow their published pricing.

Learn about pricing modes

Standard

Type	Condition	Input	Output
Text	-	¥0.0020/ 1K	¥0.0080/ 1K

Qwen3-235B-A22B

Benchmark Results

Qwen3-235B-A22B currently shows benchmark results led by GSM8K (2 / 26, score 96.40), MATH-500 (7 / 44, score 98), AIME 2024 (20 / 62, score 85.70). This page also consolidates core specs, context limits, and API pricing so you can evaluate the model from benchmark results and deployment constraints together.

General Knowledge

7 evaluations

Benchmark / mode

Score

Rank/total

BBH

88.87

8 / 21

MMLU

85.80

36 / 66

MMLU Pro

72.90

88 / 132

GPQA Diamond

71.10

114 / 187

GPQA Diamond

71.10

114 / 187

HLE

7.60

157 / 172

ARC-AGI

4.30

67 / 68

Math and Reasoning

7 evaluations

Benchmark / mode

Score

Rank/total

MATH-500

7 / 44

MATH-500

96.20

19 / 44

GSM8K

96.40

2 / 26

AIME 2024

85.70

20 / 62

AIME 2024

85.70

20 / 62

AIME2025

81.50

55 / 107

AIME2025

24.70

103 / 107

Reading Comprehension

1 evaluations

Benchmark / mode

Score

Rank/total

DROP

88.70

5 / 9

Common Sense

1 evaluations

Benchmark / mode

Score

Rank/total

SimpleQA

40 / 47

Coding and Software Engineer

3 evaluations

Benchmark / mode

Score

Rank/total

LiveCodeBench

70.70

50 / 123

LiveCodeBench

70.70

50 / 123

SWE-bench Verified

34.40

105 / 112

Writing and Creative Capabilities

2 evaluations

Benchmark / mode

Score

Rank/total

Creative Writing

84.60

11 / 23

Creative Writing

80.40

18 / 23

Common Sense Reasoning

1 evaluations

Benchmark / mode

Score

Rank/total

Simple Bench

Thinking Mode

47 / 63

Agent Level Benchmark

2 evaluations

Benchmark / mode

Score

Rank/total

Aider-Polyglot

Standard Mode

59.60

23 / 59

τ²-Bench

34.40

41 / 43

View benchmark analysis Compare with other models

Compare with other models

No curated comparisons for this model yet.

Want a custom combination? Open the compare tool

Qwen3-235B-A22B

Publisher

Alibaba

View publisher details

Qwen3-235B-A22B

Model Overview

Qwen3-235B-A22B is a reasoning model from Alibaba, released on 2025-04-28.

It accepts text input and produces text output. Its cataloged capabilities include Reasoning model and Multilingual. The cataloged parameter count is 235B, with 22B active parameters per inference. The recorded context window is 128K, and the recorded maximum output is 16K.

The checkpoint is listed under the Apache 2.0 license, with the license link included in the model references. The page records 2 API pricing rules from the listed provider; current provider pricing and conditions should be checked before deployment. The evaluation section contains 24 cataloged benchmark results with their recorded modes and scores. The page links 5 release, model-card, repository, or provider references for checking the underlying claims. Specifications, availability, and prices can change; undisclosed values are intentionally left unstated.

Qwen3-235B-A22B

FAQ

What is Qwen3-235B-A22B?

What input and output modalities does Qwen3-235B-A22B support?

The current model record lists text as input and text as output.

What are the main recorded specifications for Qwen3-235B-A22B?

The cataloged parameter count is 235B, with 22B active parameters per inference. The recorded context window is 128K, and the recorded maximum output is 16K. Fields without a source-backed value remain undisclosed.

Does Qwen3-235B-A22B have API pricing?

The page records 2 API pricing rules from the listed provider; current provider pricing and conditions should be checked before deployment.

Are benchmark results available for Qwen3-235B-A22B?

The evaluation section contains 24 cataloged benchmark results with their recorded modes and scores. Compare only results that use the same benchmark version and evaluation mode.

Is Qwen3-235B-A22B open source?

The checkpoint is listed under the Apache 2.0 license, with the license link included in the model references. Review the linked license text before commercial or derivative use.

DataLearner on WeChat

Follow DataLearner on WeChat for AI model updates and research notes.