QwenQwen 3Open SourceFeb 9, 2026
Qwen: Qwen3 Max Thinking
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 delivers major gains in factual accuracy, complex reasoning, instruction following, alignment with human preferences, and agentic behavior.
Context Window
262K
tokens
Max Output
66K
tokens
Released
Feb 9, 2026
Arena Rank
—
Capabilities
👁Vision
🧠Reasoning
🔧Tool Calling
⚡Prompt Caching
🖥Computer Use
🎨Image Generation
Supported Parameters
Include Reasoning
Show reasoning tokens
Max Tokens
Output length limit
Presence Penalty
Encourage new topics
Reasoning
Extended thinking
Response Format
JSON mode / structured output
Seed
Deterministic outputs
structured_outputs
Temperature
Controls randomness
Tool Choice
Control tool usage
Tool Calling
Function calling support
Top P
Nucleus sampling
Pricing Comparison
| Router | Input / 1M | Output / 1M | Cached Input / 1M |
|---|---|---|---|
| OpenRouter | $1.20 | $6.00 | — |
| Vercel AI | $1.20 | $6.00 | — |
Model IDs
OpenRouter
qwen/qwen3-max-thinkingTags
reasoningtool-calling
Related Models
Qwen#34
Qwen3 Max
262K ctx$0.84/1M in
Qwen#51
Qwen: Qwen3 VL 235B A22B Instruct
262K ctx$0.20/1M in
Qwen#64
Qwen: Qwen3 Next 80B A3B Instruct (free)
262K ctxFree/1M in
Qwen#67
Qwen3 235B A22B Thinking 2507
262K ctx$0.30/1M in
Qwen#69
Qwen: Qwen3 VL 235B A22B Thinking
131K ctxFree/1M in
Qwen#90
Qwen: Qwen3 235B A22B
41K ctx$0.30/1M in