Qwen3.7 Max
Qwen3.7 MaxAlibabaProprietaryReasoningVerifiedReleased May 16, 2026Input: text
Overall score
62.6
Rank 47 of 222 · 52.9–72.3
Context window
1M
1,000,000 tokens
Price per 1M tokens
$1.48 in · $4.42 out
Source: OpenRouter
Blended price
$2.21
3:1 input:output tokens
Output speed
—
Not measured yet
Category scores
Scores per capability area with the model's rank among ranked models in that category.
- 1Multilingual#1100.0
- 2Instruction following#1692.8
- 3Reasoning81.4
- 4Math#279.0
- 5Knowledge#4773.5
- 6Agentic & tool use#6873.1
- 7Coding#5665.5
Benchmark results (37)
As reported in system cards, papers and public evaluations. Position compares this model with every other model reporting the same benchmark.
Agentic & tool use
| Gert Labs | 64.27 | #5 of 52 |
| MCP Atlas | 76.4 | #15 of 42 |
| Terminal-Bench 2.0 | 69.7 | #5 of 39 |
| Claw-Eval | 65.2 | #13 of 36 |
| HLE w/ tools | 53.5 | #17 of 27 |
| BFCL v4 | 75.0 | #2 of 19 |
| ResearchClawBench | 18.7 | #6 of 19 |
| VITA-Bench | 47.9 | #1 of 11 |
| QwenClawBench | 64.3 | #1 of 10 |
| QwenWebBench | 1568 | #1 of 5 |
Coding
| SWE-bench Verified | 80.4 | #16 of 98 |
| SWE-bench Pro | 60.6 | #26 of 69 |
| LiveCodeBench v6 | 91.6 | #4 of 46 |
| NL2Repo | 47.2 | #15 of 31 |
| SciCode | 53.5 | #2 of 23 |
| OpenHarmony Bench | 53.4 | #9 of 12 |
Multilingual
| SWE Multilingual | 78.3 | #15 of 46 |
| MMLU-ProX | 87.0 | #1 of 16 |
| INCLUDE | 86.2 | #3 of 10 |
| NOVA-63 | 59.0 | #2 of 7 |
| PolyMath | 86.5 | #1 of 6 |
| MAXIFE | 89.2 | #1 of 5 |
Knowledge
| GPQA | 92.4 | #12 of 117 |
| HLE | 41.4 | #29 of 69 |
| MMLU-Pro | 89.6 | #1 of 67 |
| SuperGPQA | 73.6 | #4 of 20 |
| MMLU-Redux | 95.0 | #2 of 15 |
| MMMLU | 90.3 | #2 of 11 |
Math
| FrontierMath v2 (Tiers 1-3) | 64.561 | #31 of 64 |
| FrontierMath v2 (Tier 4) | 34.146 | #23 of 51 |
| HMMT Feb 2026 | 97.1 | #1 of 24 |
| IMOAnswerBench | 90.0 | #2 of 9 |
| Apex | 44.5 | #4 of 8 |
Instruction following
| IFEval | 94.3 | #5 of 53 |
| IFBench | 79.1 | #14 of 42 |
Reasoning
| MRCRv2 | 90.4 | #3 of 12 |
Multimodal
| Design Arena Website | 1274 | #24 of 91 |
Pricing by source
USD per 1M tokens.
| Source | Input | Output | Cached in |
|---|---|---|---|
| List price (OpenRouter) | $1.48 | $4.42 | — |
| OpenRouter qwen/qwen3.7-max | $1.48 | $4.42 | $0.29 |
Qwen's May 16, 2026 launch post says qwen3.7-max will be available soon through Alibaba Cloud Model Studio and shows API usage, but does not publish exact token pricing.
Arena ratings
Human-preference ratings from LMArena. Best-rated variant shown.
| Text(preview) | 1475 | #24 · 3.9K votes |
| WebDev (coding)(20260517) | 1516 | #57 · 8.9K votes |
Speed
Measured by us through OpenRouter (median of trials).
Not measured yet.
Related models
Same family
Similar score
- Muse Spark Open62.8 · $2.00
- Gemini 3.6 Flash62.8 · $1.50
- Qwen3.8-Flash-Next62.9 · —
- Claude Opus 4.662.3 · $10.00
- Ornith-1.5-397B62.3 · —