Home/Models/Qwen3 235B A22B Instruct 2507
Alibaba Qwen logo

Qwen3 235B A22B Instruct 2507

by Alibaba Qwen · Released Jul 2025

Open Source
45.7
avg score
Rank #121
Compare
Better than 48% of all models
Context
262K tokens (~131 books)
Input $/1M
$0.07
Output $/1M
$0.10
Type
text
License
Open Source
Benchmarks
20 tested
Data updated today
About

Qwen3-235B-A22B-Instruct-2507 is a multilingual, instruction-tuned mixture-of-experts language model based on the Qwen3-235B architecture, with 22B active parameters per forward pass. It is optimized for general-purpose text generation, including instruction following,...

Tested on 20 benchmarks with 48.5% average. Top scores: Chatbot Arena Elo — Overall (1422.6%), OpenCompass — IFEval (88.3%), OpenCompass — MMLU-Pro (79.2%).

Capabilities
coding
44.8
#84 globally
reasoning
28.9
#86 globally
math
68.8
#35 globally
knowledge
53.7
#81 globally
language
58.7
#89 globally
Benchmark Scores
Compare All
Tested on 20 benchmarks · Ranked across 6 categories
Score Distribution (all 233 models)
0255075100
▲ You are here
LiveBench — Coding

Regularly refreshed coding problems that avoid data contamination. New problems added monthly to prevent memorization.

69.6
Aider polyglot

Multi-language code editing from Aider. Tests editing ability across Python, JavaScript, TypeScript, Java, C++, Go, Rust, and more.

59.6
OpenCompass — LiveCodeBenchV6

OpenCompass Live Code Bench v6. Fresh competitive programming problems to evaluate code generation without memorization.

43.0
LiveBench — Reasoning

Regularly refreshed reasoning problems testing logical deduction, spatial reasoning, and analytical thinking.

58.4
LiveBench — Data Analysis

Fresh data analysis tasks testing ability to interpret tables, charts, and statistical data.

44.7
ARC-AGI

Abstraction and Reasoning Corpus. Tests fluid intelligence through novel visual pattern recognition puzzles. Core measure of general intelligence.

11.0
OpenCompass — AIME2025

OpenCompass evaluation on AIME 2025 problems. Tests mathematical reasoning on fresh competition problems.

69.5
LiveBench — Mathematics

Regularly updated math problems that test numerical reasoning, algebra, calculus, and combinatorics.

68.0
Excellent (85+) Good (70-85) Average (50-70) Below (<50)
Model Family · Alibaba Qwen Qwen 3
See the full Qwen 3 family →
Links
Documentation
Community
BenchGecko API
qwen3-235b-a22b-2507
Specifications
  • Typetext
  • Context262K tokens (~131 books)
  • ReleasedJul 2025
  • LicenseOpen Source
  • StatusActive
  • Cost / Message~$0.000
Available On
Alibaba Qwen logoAlibaba Qwen$0.07
Share & Export
Tweet
Qwen3 235B A22B Instruct 2507 is an open-source text AI model by Alibaba Qwen, released in July 2025. It has an average benchmark score of 45.7. Context window: 262K tokens.