Qwen3-Max: pricing, context window and benchmarks
Qwen3-Max is a frontier model from Alibaba Cloud. It offers a 400,000-token context window, costs $17.37 per million input tokens and $94.30 per million output tokens, and averages 84.4 across the index's benchmark suite.
Pricing
Qwen3-Max costs $17.37 per million input tokens and $94.30 per million output tokens. A blended 3:1 input-to-output mix works out to about $36.60 per million tokens. Cached input is listed at $2.02 per million tokens.
| Item | Price | Notes |
|---|---|---|
| Input | $17.37 | per 1M tokens |
| Output | $94.30 | per 1M tokens |
| Cached input | $2.02 | per 1M tokens |
| Blended 3:1 | $36.60 | estimate per 1M tokens |
| 1M in + 1M out | $111.67 | estimate |
Benchmarks
Scores below are modelled estimates on a 0–100 scale. Use them to compare relative position, not as official vendor results.
| Benchmark | Score | Rank | Percentile | Relative |
|---|---|---|---|---|
| MMLU-Pro | 84.0 | 83 | 86th | |
| GPQA Diamond | 82.3 | 26 | 96th | |
| SWE-bench Verified | 73.3 | 62 | 90th | |
| HumanEval | 88.5 | 66 | 89th | |
| MATH-500 | 89.7 | 53 | 91th | |
| AIME 2025 | 81.1 | 74 | 88th | |
| LiveCodeBench | 82.2 | 24 | 96th | |
| IFEval | 92.5 | 31 | 95th | |
| TruthfulQA | 83.4 | 99 | 84th | |
| HellaSwag | 92.9 | 25 | 96th | |
| MMMU | 78.7 | 81 | 87th | |
| Arena Elo | 1449.2 | 74 | 88th |
What Qwen3-Max is good for
Qwen3-Max sits in the frontier tier of the index. It is aimed at work where answer quality matters more than unit cost: hard reasoning, multi-step coding, document analysis and multimodal inputs. At $17.37 per million input tokens and a 400,000-token context window, it is a premium option, so reserve it for tasks where a cheaper model measurably fails.
Strengths
- Top-tier average benchmark score (84.4), placing it #55 of 600.
- Large 400,000-token context window for long documents and agent traces.
- Accepts more than text, including image input where the provider supports it.
- Reasoning-oriented profile, useful for multi-step problems and structured analysis.
- Coding benchmarks are a relative strength within its tier.
Trade-offs
- Proprietary licence: you cannot download the weights or self-host the model.
- Premium input pricing at $17.37 per million tokens.
- Cost can climb quickly on long outputs and high-volume traffic.
How it compares
| Model | Provider | Context | Input / 1M | Avg score |
|---|---|---|---|---|
| GPT-4.1 Max | OpenAI | 400000 | $17.34 | 84.7 |
| Qwen3-Max Turbo | Alibaba Cloud | 1000000 | $17.70 | 84.7 |
| o3 | OpenAI | 400000 | $16.80 | 84.5 |
| o4 Vision | OpenAI | 200000 | $17.01 | 84.5 |
| Qwen3-Max | Alibaba Cloud | 400000 | $17.37 | 84.4 |
| Gemini 3 Flash Turbo | Google DeepMind | 200000 | $17.07 | 84.1 |
FAQ
How much does Qwen3-Max cost?
Qwen3-Max is listed at $17.37 per million input tokens and $94.30 per million output tokens. Cached input is $2.02 per million tokens. A blended 3:1 input-to-output mix is about $36.60 per million tokens.
What is the context window of Qwen3-Max?
Qwen3-Max has a context window of 400,000 tokens and can emit up to 4,096 output tokens per request in this index.
Is Qwen3-Max open weights?
No. Qwen3-Max is a proprietary model with a proprietary licence, so you cannot download the weights.
How does Qwen3-Max score on benchmarks?
Qwen3-Max averages 84.4 across the scored benchmarks in this index, ranking #55 of 600 models. Its strongest relative result is HellaSwag at 92.9.
What is Qwen3-Max best for?
Qwen3-Max sits in the frontier tier of the index. It is aimed at work where answer quality matters more than unit cost: hard reasoning, multi-step coding, document analysis and multimodal inputs. At $17.37 per million input tokens and a 400,000-token context window, it is a premium option, so reserve it for tasks where a cheaper model measurably fails.