Qwen3 235B A22B Thinking 2507
by Qwentext->text3 endpoints
Neura Intelligence Index
58.1/ 100
Rank
#146—
Confidence
mediumBenchmarks
4(10% coverage)
Overview
Qwen3-235B-A22B-Thinking-2507 is a high-performance, open-weight Mixture-of-Experts (MoE) language model optimized for complex reasoning tasks. It activates 22B of its 235B parameters per forward pass and natively supports up to 262,144...
Capabilities
Text generation
Tool use / Function calling
Structured output
Extended reasoning
Long context
Top-P sampling
Stop sequences
Deterministic seed
Log probabilities
Modalities
Input
Text
Output
Text
Technical Specifications
Context Window
131.1K tokens
Max Output
118.0K tokens
Knowledge Cutoff
2025-06-30
Tokenizer
Qwen3
Uptime (24h)
10000.0%
Parameters
235000000000B
License
apache_2_0Open Weights
Country
🇨🇳CN
Release Date
July 25, 20251y
Supported Parameters
frequency_penaltyinclude_reasoninglogprobsmax_tokenspresence_penaltyreasoningrepetition_penaltyresponse_formatseedstoptemperaturetool_choicetoolstop_ktop_logprobstop_p
Benchmark Performance
math
| Benchmark | Score | Source |
|---|---|---|
| AIME 2025 | 92.3 | verified |
reasoning
| Benchmark | Score | Source |
|---|---|---|
| GPQA Diamond | 81.1 | verified |
| Humanity's Last Exam | 18.2 | verified |
agent
| Benchmark | Score | Source |
|---|---|---|
| TAU-Bench Retail | 67.8 | verified |
Performance Over Time
Pricing
Per 1M tokens
Input$0.23
Output$2.30
Added
July 25, 2025
Last synced 9/18/2026