Qwen3-Max Max: pricing, context window and benchmarks
Qwen3-Max Max is a frontier model from Alibaba Cloud. It offers a 400,000-token context window, costs $18.53 per million input tokens and $74.05 per million output tokens, and averages 85.9 across the index's benchmark suite.
Pricing
Qwen3-Max Max costs $18.53 per million input tokens and $74.05 per million output tokens. A blended 3:1 input-to-output mix works out to about $32.41 per million tokens. Cached input is listed at $2.38 per million tokens.
| Item | Price | Notes |
|---|---|---|
| Input | $18.53 | per 1M tokens |
| Output | $74.05 | per 1M tokens |
| Cached input | $2.38 | per 1M tokens |
| Blended 3:1 | $32.41 | estimate per 1M tokens |
| 1M in + 1M out | $92.58 | 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 | 85.2 | 62 | 90th | |
| GPQA Diamond | 83.8 | 14 | 98th | |
| SWE-bench Verified | 74.3 | 49 | 92th | |
| HumanEval | 90.2 | 44 | 93th | |
| MATH-500 | 92.8 | 14 | 98th | |
| AIME 2025 | 82.7 | 58 | 90th | |
| LiveCodeBench | 82.8 | 16 | 97th | |
| IFEval | 93.7 | 15 | 98th | |
| TruthfulQA | 89.3 | 8 | 99th | |
| HellaSwag | 91.4 | 51 | 92th | |
| MMMU | 78.9 | 76 | 87th | |
| Arena Elo | 1467.7 | 38 | 94th |
What Qwen3-Max Max is good for
Qwen3-Max 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 $18.53 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 (85.9), placing it #37 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 $18.53 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-5.1 | OpenAI | 2000000 | $19.01 | 86.0 |
| Claude Opus 4.5 Vision | Anthropic | 200000 | $17.70 | 85.9 |
| Claude Sonnet 4.5 Turbo | Anthropic | 1000000 | $18.37 | 85.9 |
| Grok 3 Max | xAI | 2000000 | $17.66 | 85.9 |
| Qwen3-Max Max | Alibaba Cloud | 400000 | $18.53 | 85.9 |
| Qwen3-Max Reasoning | Alibaba Cloud | 2000000 | $18.12 | 85.9 |
FAQ
How much does Qwen3-Max Max cost?
Qwen3-Max Max is listed at $18.53 per million input tokens and $74.05 per million output tokens. Cached input is $2.38 per million tokens. A blended 3:1 input-to-output mix is about $32.41 per million tokens.
What is the context window of Qwen3-Max Max?
Qwen3-Max Max has a context window of 400,000 tokens and can emit up to 65,536 output tokens per request in this index.
Is Qwen3-Max Max open weights?
No. Qwen3-Max Max is a proprietary model with a proprietary licence, so you cannot download the weights.
How does Qwen3-Max Max score on benchmarks?
Qwen3-Max Max averages 85.9 across the scored benchmarks in this index, ranking #37 of 600 models. Its strongest relative result is TruthfulQA at 89.3.
What is Qwen3-Max Max best for?
Qwen3-Max 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 $18.53 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.