Qwen3-Max Turbo: pricing, context window and benchmarks
Qwen3-Max Turbo is a frontier model from Alibaba Cloud. It offers a 1,000,000-token context window, costs $17.70 per million input tokens and $63.95 per million output tokens, and averages 84.7 across the index's benchmark suite.
Pricing
Qwen3-Max Turbo costs $17.70 per million input tokens and $63.95 per million output tokens. A blended 3:1 input-to-output mix works out to about $29.26 per million tokens. Cached input is listed at $2.61 per million tokens.
| Item | Price | Notes |
|---|---|---|
| Input | $17.70 | per 1M tokens |
| Output | $63.95 | per 1M tokens |
| Cached input | $2.61 | per 1M tokens |
| Blended 3:1 | $29.26 | estimate per 1M tokens |
| 1M in + 1M out | $81.65 | 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 | 86.8 | 41 | 93th | |
| GPQA Diamond | 78.3 | 79 | 87th | |
| SWE-bench Verified | 75.9 | 31 | 95th | |
| HumanEval | 92.5 | 15 | 98th | |
| MATH-500 | 91.9 | 24 | 96th | |
| AIME 2025 | 81.2 | 71 | 88th | |
| LiveCodeBench | 76.8 | 72 | 88th | |
| IFEval | 92.1 | 38 | 94th | |
| TruthfulQA | 86.3 | 46 | 92th | |
| HellaSwag | 90.8 | 67 | 89th | |
| MMMU | 78.8 | 79 | 87th | |
| Arena Elo | 1456.9 | 59 | 90th |
What Qwen3-Max Turbo is good for
Qwen3-Max Turbo 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 $17.70 per million input tokens and a 1,000,000-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 (84.7), placing it #52 of 600.
- Large 1,000,000-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 $17.70 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 Vision | OpenAI | 400000 | $17.76 | 84.8 |
| Gemini 2.5 Pro Max | Google DeepMind | 200000 | $16.94 | 84.8 |
| Gemini 3 Flash Pro | Google DeepMind | 400000 | $17.71 | 84.8 |
| o3 Max | OpenAI | 2000000 | $17.82 | 84.8 |
| GPT-4.1 Max | OpenAI | 400000 | $17.34 | 84.7 |
| Qwen3-Max Turbo | Alibaba Cloud | 1000000 | $17.70 | 84.7 |
FAQ
How much does Qwen3-Max Turbo cost?
Qwen3-Max Turbo is listed at $17.70 per million input tokens and $63.95 per million output tokens. Cached input is $2.61 per million tokens. A blended 3:1 input-to-output mix is about $29.26 per million tokens.
What is the context window of Qwen3-Max Turbo?
Qwen3-Max Turbo has a context window of 1,000,000 tokens and can emit up to 128,000 output tokens per request in this index.
Is Qwen3-Max Turbo open weights?
No. Qwen3-Max Turbo is a proprietary model with a proprietary licence, so you cannot download the weights.
How does Qwen3-Max Turbo score on benchmarks?
Qwen3-Max Turbo averages 84.7 across the scored benchmarks in this index, ranking #52 of 600 models. Its strongest relative result is Arena Elo at 1456.9.
What is Qwen3-Max Turbo best for?
Qwen3-Max Turbo 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 $17.70 per million input tokens and a 1,000,000-token context window, it is a premium option, so reserve it for tasks where a cheaper model measurably fails.