Qwen3.7 MaxvsQwen3.6-Max-Preview
Across 11 shared benchmarks, Qwen3.7 Max leads overall: Qwen3.7 Max wins 11, Qwen3.6-Max-Preview wins 0, with 0 ties and an average score difference of +3.92.
Qwen3.7 Max
阿里巴巴 · 2026-05-20 · Reasoning model
Qwen3.6-Max-Preview
阿里巴巴 · 2026-04-18 · Chat model
Qwen3.7 Max11 wins(100%)(0%)0 winsQwen3.6-Max-Preview
Benchmark scores
Grouped by capability, sorted by largest gap within each. 11 shared benchmarks.
Coding and Software Engineer
Qwen3.7 Max 4/4| Benchmark | Qwen3.7 Max | Qwen3.6-Max-Preview | Diff |
|---|---|---|---|
| LiveCodeBench | 91.604 / 126最高（无工具） | 87.1012 / 126最高（无工具） | +4.50 |
| SWE-bench Multilingual | 78.304 / 25Thinking (With Tools) | 73.8010 / 25Thinking (With Tools) | +4.50 |
| SWE-Bench Pro - Public | 60.6012 / 57Thinking (With Tools) | 57.3020 / 57Deep Thinking (With Tools) | +3.30 |
| SWE-bench Verified | 80.4013 / 114Thinking (With Tools) | 78.8021 / 114Thinking (With Tools) | +1.60 |
General Knowledge
Qwen3.7 Max 2/2| Benchmark | Qwen3.7 Max | Qwen3.6-Max-Preview | Diff |
|---|---|---|---|
| HLE | 53.5018 / 181Thinking (With Tools) | 50.2029 / 181Thinking (With Tools) | +3.30 |
| MMLU Pro | 89.604 / 133最高（无工具） | 88.505 / 133最高（无工具） | +1.10 |
AI Agent - Tool Usage
Qwen3.7 Max 1/1| Benchmark | Qwen3.7 Max | Qwen3.6-Max-Preview | Diff |
|---|---|---|---|
| Terminal Bench 2.0 | 69.705 / 48Thinking (With Tools) | 65.4011 / 48Deep Thinking (With Tools) | +4.30 |
Commonsense Reasoning
Qwen3.7 Max 1/1| Benchmark | Qwen3.7 Max | Qwen3.6-Max-Preview | Diff |
|---|---|---|---|
| SimpleBench | 70.407 / 67Normal (No Tools) | 6311 / 67Normal (No Tools) | +7.40 |
General Evaluation
Qwen3.7 Max 1/1| Benchmark | Qwen3.7 Max | Qwen3.6-Max-Preview | Diff |
|---|---|---|---|
| GPQA Diamond | 92.4022 / 226最高（无工具） | 90.4036 / 226最高（无工具） | +2 |
Instruction Following
Qwen3.7 Max 1/1| Benchmark | Qwen3.7 Max | Qwen3.6-Max-Preview | Diff |
|---|---|---|---|
| IF Bench | 79.104 / 33最高（无工具） | 74.209 / 33最高（无工具） | +4.90 |
Math and Reasoning
Qwen3.7 Max 1/1| Benchmark | Qwen3.7 Max | Qwen3.6-Max-Preview | Diff |
|---|---|---|---|
| IMO-AnswerBench | 903 / 23最高（无工具） | 83.8014 / 23最高（无工具） | +6.20 |
Specs
| Field | Qwen3.7 Max | Qwen3.6-Max-Preview |
|---|---|---|
| Publisher | 阿里巴巴 | 阿里巴巴 |
| Release date | 2026-05-20 | 2026-04-18 |
| Model type | Reasoning model | Chat model |
| Architecture | Dense | Dense |
| Parameters | Not available | Not available |
| Context length | 1M | 262K |
| Max output | 64K | 64K |
API pricing
Prices use DataLearner records when available; missing fields are not inferred.
| Item | Qwen3.7 Max | Qwen3.6-Max-Preview |
|---|---|---|
| Text input | ¥12 / 1M tokens | $1.3 / 1M tokens |
| Text output | ¥36 / 1M tokens | $7.8 / 1M tokens |
Summary
- Qwen3.7 Maxleads in:Coding and Software Engineer (4/4), General Knowledge (2/2), AI Agent - Tool Usage (1/1), Commonsense Reasoning (1/1), General Evaluation (1/1), Instruction Following (1/1), Math and Reasoning (1/1)
On average across the 11 shared benchmarks, Qwen3.7 Max scores 3.92 higher.
Largest single-benchmark gap: SimpleBench — Qwen3.7 Max 70.40 vs Qwen3.6-Max-Preview 63 (+7.40).
Page generated from structured model, pricing and benchmark records. No real-time LLM is used to write the prose.