Alibaba

Qwen3-235B-A22B-Instruct-2507

Zero-eval
#1ZebraLogic
#1MultiPL-E
#1Creative Writing v3
+19 more

by Alibaba

About

Qwen3-235B-A22B-Instruct-2507 is a language model developed by Alibaba. It achieves strong performance with an average score of 72.1% across 25 benchmarks. It excels particularly in ZebraLogic (95.0%), MMLU-Redux (93.1%), IFEval (88.7%). It's licensed for commercial use, making it suitable for enterprise applications. Released in 2025, it represents Alibaba's latest advancement in AI technology.

Timeline
AnnouncedJul 22, 2025
ReleasedJul 22, 2025
Specifications
License & Family
License
Apache 2.0
Benchmark Performance Overview
Performance metrics and category breakdown

Overall Performance

25 benchmarks
Average Score
72.1%
Best Score
95.0%
High Performers (80%+)
8

Top Categories

roleplay
75.4%
general
73.7%
code
70.3%
reasoning
68.4%
math
50.2%
Benchmark Performance
Top benchmark scores with normalized values (0-100%)
Ranking Across Benchmarks
Position relative to other models on each benchmark

ZebraLogic

Rank #1 of 2
#1Qwen3-235B-A22B-Instruct-2507
95.0%
#2Kimi K2 Instruct
89.0%

MMLU-Redux

Rank #2 of 13
#1DeepSeek-R1-0528
93.4%
#2Qwen3-235B-A22B-Instruct-2507
93.1%
#3DeepSeek-R1
92.9%
#4Kimi K2 Instruct
92.7%
#5DeepSeek-V3
89.1%

IFEval

Rank #11 of 37
#8Gemma 3 12B
88.9%
#9Llama 3.1 Nemotron Ultra 253B v1
89.5%
#10Nova Lite
89.7%
#11Qwen3-235B-A22B-Instruct-2507
88.7%
#12Llama 3.1 405B Instruct
88.6%
#13GPT-4.5
88.2%
#14Llama 3.1 70B Instruct
87.5%

MultiPL-E

Rank #1 of 10
#1Qwen3-235B-A22B-Instruct-2507
87.9%
#2Kimi K2 Instruct
85.7%
#3Qwen2.5 32B Instruct
75.4%
#4Qwen2.5 72B Instruct
75.1%

Creative Writing v3

Rank #1 of 1
#1Qwen3-235B-A22B-Instruct-2507
87.5%
All Benchmark Results for Qwen3-235B-A22B-Instruct-2507
Complete list of benchmark scores with detailed information
ZebraLogic
ZebraLogic benchmark
reasoning
text
0.95
95.0%
Self-reported
MMLU-Redux
MMLU-Redux benchmark
general
text
0.93
93.1%
Self-reported
IFEval
IFEval benchmark
code
text
0.89
88.7%
Self-reported
MultiPL-E
MultiPL-E benchmark
general
text
87.90
87.9%
Self-reported
Creative Writing v3
Creative Writing v3 benchmark
general
text
0.88
87.5%
Self-reported
WritingBench
WritingBench benchmark
general
text
0.85
85.2%
Self-reported
CSimpleQA
CSimpleQA benchmark
general
text
0.84
84.3%
Self-reported
MMLU-Pro
MMLU-Pro benchmark
general
text
0.83
83.0%
Self-reported
Include
Include benchmark
general
text
0.80
79.5%
Self-reported
MMLU-ProX
MMLU-ProX benchmark
general
text
0.79
79.4%
Self-reported
Showing 1 to 10 of 25 benchmarks