Qwen3 Max Thinking
by Qwentext->text1 endpoint
Neura Intelligence Index
66.9/ 100
Rank
#103—
Confidence
mediumBenchmarks
5(13% coverage)
Overview
Qwen3-Max-Thinking is the flagship reasoning model in the Qwen3 series, designed for high-stakes cognitive tasks that require deep, multi-step reasoning. By significantly scaling model capacity and reinforcement learning compute, it...
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
262.1K tokens
Max Output
65.5K tokens
Tokenizer
Qwen
Uptime (24h)
10000.0%
Parameters
1000000000000B
Country
🇨🇳CN
Release Date
February 13, 20267mo
Supported Parameters
frequency_penaltyinclude_reasoninglogprobsmax_tokenspresence_penaltyreasoningresponse_formatseedstopstructured_outputstemperaturetool_choicetoolstop_ktop_logprobstop_p
Benchmark Performance
general
| Benchmark | Score | Source |
|---|---|---|
| MMMLU | 84.4 | verified |
coding
| Benchmark | Score | Source |
|---|---|---|
| SWE-bench Verified | 75.3 | verified |
reasoning
| Benchmark | Score | Source |
|---|---|---|
| GPQA Diamond | 87.4 | verified |
agent
| Benchmark | Score | Source |
|---|---|---|
| BrowseComp | 53.9 | verified |
| Toolathlon | 18.8 | verified |