Qwen: Qwen3 Max
Qwen3-Max is an updated release built on the Qwen3 series, offering major improvements in reasoning, instruction following, multilingual support, and long-tail knowledge coverage compared to the January 2025 version. It delivers higher accuracy in math, coding, logic, and science tasks, follows complex instructions in Chinese and English more reliably, reduces hallucinations, and produces higher-quality responses for open-ended Q&A, writing, and conversation. The model supports over 100 languages with stronger translation and commonsense reasoning, and is optimized for retrieval-augmented generation (RAG) and tool calling, though it does not include a dedicated “thinking” mode.
Pricing per 1M Tokens
| Input (Prompt) | $0.78 |
| Output (Completion) | $3.90 |
| Cache Read | $0.16 |
| Cache Write | Free |
| Image | N/A |
Specifications
| Context Length | 262K |
| Max Output Tokens | 33K |
| Input Modalities | Text |
| Output Modalities | Text |
| Tokenizer | Qwen3 |
| Instruct Type | N/A |
| Top Provider Context | 262K |
| Top Provider Max Output | 33K |
| Moderated | No |
Compare this model
See how Qwen: Qwen3 Max stacks up against other models.
More from Qwen
Last updated: March 23, 2026
First tracked: March 23, 2026