Qwen3 Max. The latest Qwen 3 model (5 september 2025). Higher accuracy in coding and science, better instruction following, and optimized for tool calling.
Added Sep 5, 2025
Context Window
256.0K
Max Output
32.8K
Input Price (Auto)
$1.20/1M
Output Price (Auto)
$6.00/1M
Cache Read (Auto)
$0.60/1M
Benchmarks
Benchmarks
Performance metrics and benchmarks
Sourced from Artificial Analysis.
Intelligence Index
24.0
Reasoning
GPQA Diamond
Graduate-level scientific reasoning
76.4%
Better than 68% of models compared
HLE
Humanity's Last Exam
11.1%
Better than 69% of models compared
IFBench
Instruction-following benchmark
44.1%
Better than 50% of models compared
T²-Bench Telecom
Conversational AI agents in dual-control scenarios
74.3%
Better than 67% of models compared
AA-LCR
Long context reasoning evaluation
46.7%
Better than 59% of models compared
Coding
SciCode
Python programming for scientific computing
38.3%
Better than 70% of models compared
Terminal-Bench Hard
Agentic coding and terminal use
20.5%
Better than 61% of models compared
LiveCodeBench
Contamination-free coding benchmark
76.7%
Better than 88% of models compared
Math
AIME 2025
American Invitational Mathematics Examination 2025
80.7%
Better than 77% of models compared
Knowledge
MMLU-Pro
Professional and academic subject knowledge
84.1%
Better than 88% of models compared
Last updated Jun 28, 2026
Artificial AnalysisProviders
Auto routing is available for this model. Explicit provider selection is not available.
Loading provider options…
Related text models
Compare Qwen3 Max with similar models from the same provider or model family.
Qwen3.6 35B A3B Thinking
Qwen/Qwen3.6-35B-A3B:thinkingQwen3.6 35B A3B is a native vision-language MoE model with hybrid attention. Compared to Qwen3.5 35B A3B, Alibaba reports stronger agentic coding, mathematical and code reasoning, and better spatial understanding (including object localization and detection).
Qwen3.6 35B A3B
Qwen/Qwen3.6-35B-A3BQwen3.6 35B A3B is a native vision-language MoE model with hybrid attention. Compared to Qwen3.5 35B A3B, Alibaba reports stronger agentic coding, mathematical and code reasoning, and better spatial understanding (including object localization and detection).
Qwen 3 Coder 480B
qwen/qwen3-coderQwen 3 Coder 480B, a 480 billion total parameter model with 35B active, and 160 total experts with 8 active. Performs similar to Claude 4 Sonnet in coding benchmarks, but does so at a much lower price.
Qwen3.5 9B
qwen/qwen3.5-9bQwen3.5 9B is a multimodal foundation model from the Qwen 3.5 family, built for efficient reasoning, coding, and visual understanding in a compact 9B architecture.
Qwen3.5 397B A17B
qwen/qwen3.5-397b-a17bQwen 3.5's open-source 397B MoE model (17B active params) with hybrid linear attention. Supports text, image, and video input with a 256K context window.
Qwen3.5 397B A17B Thinking
qwen/qwen3.5-397b-a17b-thinkingQwen 3.5's open-source 397B MoE model (17B active params) with hybrid linear attention and extended reasoning. Supports text, image, and video input with a 256K context window.