Models · Qwen (Alibaba) · Out sinceReleased 16 May 2026
Qwen3.7 Max
Available on OpenRouterOn OpenRouterAvailable nowGenerally availableClosed (can't be downloaded)Proprietary
Qwen3.7 Max is made by Qwen (Alibaba). We don't have enough test results yet to rank it. It's mid-priced to use.
Previous-gen proprietary flagship; text-only. Thinking toggled via enable_thinking; preserve_thinking recommended for agentic tasks. Vendor recommends system prompt 'Reasoning effort is set to xhigh...' for reasoning scenarios.
- —
- —
- —
- Mid-priced$2.50 / $7.50
- $3.75
- 1M
- 131K
- 33 (8 independent8 indep.)
- 16 May 2026
- Qwen (Alibaba)
- text
Scores are out of 100 for each area, compared with every model we track; “#” is its rank. Context is how much text it can read at once (1M is roughly 700,000 words). Blended price mixes the cost of what you send and what it writes back.
No reasoningDefault
OpenRouter is a service that gives access to many AI models in one place. This model is listed there as qwen/qwen3.7-max.
Route it as qwen/qwen3.7-max at $1.48 in / $4.42 out per 1M tokens, 1M context. Listed since 21 May 2026.
frequency_penaltyinclude_reasoninglogprobsmax_tokenspresence_penaltyreasoningresponse_formatseedstopstructured_outputstemperaturetool_choicetoolstop_ktop_logprobstop_p
Every test result
Every result
All the numbers,with where they came from.
Every result we've found for this model, with who measured it and a link to where we read it.
All raw rows (vendor and independent kept separate), with setting label, source, date, harness and notes.
| NotesNotes | ||||||
|---|---|---|---|---|---|---|
| Agents' Last Exam | 11.8% | Defaultthinking (setting not stated) | Maker's own figureVendor-reportedQwen (Alibaba) ↗ | 3 Aug 2026 | — | Qwen3.7-Max column in the Qwen3.8-Max launch table |
| AutomationBench | 14.2% | Defaultthinking (setting not stated) | Maker's own figureVendor-reportedQwen (Alibaba) ↗ | 3 Aug 2026 | — | Qwen3.7-Max column in the Qwen3.8-Max launch table |
| DeepSWE | 21.6% | Defaultthinking (setting not stated) | Maker's own figureVendor-reportedQwen (Alibaba) ↗ | 3 Aug 2026 | Claude Code / mini-SWE-agent (best) | Qwen3.7-Max column in the Qwen3.8-Max launch table |
| FrontierSWE | 40.7% | Defaultthinking (setting not stated) | Maker's own figureVendor-reportedQwen (Alibaba) ↗ | 3 Aug 2026 | Claude Code | Qwen3.7-Max column in the Qwen3.8-Max launch table |
| GPQA Diamond | 90.2% | Default | Independent testIndependentVals.ai ↗ | 1 Sep 2026 | — | ±1.497 stderr; $0.061/test |
| GPQA Diamond | 92.4% | Defaultthinking (vendor recommends 'Reasoning effort is set to xhigh' system prompt for reasoning tasks) | Maker's own figureVendor-reportedQwen (Alibaba) ↗ | 16 May 2026 | — | |
| HMMT February 2026 | 97.1% | Defaultthinking (vendor recommends 'Reasoning effort is set to xhigh' system prompt for reasoning tasks) | Maker's own figureVendor-reportedQwen (Alibaba) ↗ | 16 May 2026 | — | HMMT 2026 Feb |
| Humanity's Last Exam | 41.4% | Defaultthinking (vendor recommends 'Reasoning effort is set to xhigh' system prompt for reasoning tasks) | Maker's own figureVendor-reportedQwen (Alibaba) ↗ | 16 May 2026 | — | |
| Humanity's Last Exam (with tools) | 53.5% | Defaultthinking (vendor recommends 'Reasoning effort is set to xhigh' system prompt for reasoning tasks) | Maker's own figureVendor-reportedQwen (Alibaba) ↗ | 16 May 2026 | — | |
| IMO-AnswerBench | 90.0% | Defaultthinking (vendor recommends 'Reasoning effort is set to xhigh' system prompt for reasoning tasks) | Maker's own figureVendor-reportedQwen (Alibaba) ↗ | 16 May 2026 | — | |
| LiveCodeBench | 87.1% | Default | Independent testIndependentVals.ai ↗ | 1 Sep 2026 | — | ±0.974 stderr; $0.087/test |
| LiveCodeBench | 91.6% | Defaultthinking (vendor recommends 'Reasoning effort is set to xhigh' system prompt for reasoning tasks) | Maker's own figureVendor-reportedQwen (Alibaba) ↗ | 16 May 2026 | — | |
| LLM Creative Story-Writing Benchmark (Lech Mazur) | -2.6 | DefaultQwen 3.7 Max | Independent testIndependentLech Mazur (LLM Creative Story-Writing Benchmark) ↗ | 28 Sep 2026 | — | rank 50/56; Thurstone comparison score (centered at 0); est. win chance 15%; 95% bootstrap -2.783 to -2.523 |
| LMArena Text - Coding category | 1524 | Defaultqwen3.7-max-preview | Independent testIndependentLMArena ↗ | 30 Sep 2026 | — | text arena coding category, style control on; rank 25 (CI rank 5-77); 95% CI 1506-1542; 1167 votes |
| MathArena Apex | 44.5% | Defaultthinking (vendor recommends 'Reasoning effort is set to xhigh' system prompt for reasoning tasks) | Maker's own figureVendor-reportedQwen (Alibaba) ↗ | 16 May 2026 | — | Reported as 'Apex' |
| MCP Atlas | 76.4% | Defaultthinking (vendor recommends 'Reasoning effort is set to xhigh' system prompt for reasoning tasks) | Maker's own figureVendor-reportedQwen (Alibaba) ↗ | 16 May 2026 | — | Public set, gemini-2.5-pro judge |
| MCPMark | 60.8% | Defaultthinking (vendor recommends 'Reasoning effort is set to xhigh' system prompt for reasoning tasks) | Maker's own figureVendor-reportedQwen (Alibaba) ↗ | 16 May 2026 | — | |
| MLS-Bench-Lite | 31.7% | Defaultthinking (setting not stated) | Maker's own figureVendor-reportedQwen (Alibaba) ↗ | 3 Aug 2026 | Claude Code | Qwen3.7-Max column in the Qwen3.8-Max launch table |
| MMLU-Pro | 89.6% | Defaultthinking (vendor recommends 'Reasoning effort is set to xhigh' system prompt for reasoning tasks) | Maker's own figureVendor-reportedQwen (Alibaba) ↗ | 16 May 2026 | — | |
| MMMLU | 90.3% | Defaultthinking (vendor recommends 'Reasoning effort is set to xhigh' system prompt for reasoning tasks) | Maker's own figureVendor-reportedQwen (Alibaba) ↗ | 16 May 2026 | — | |
| NL2Repo-Bench | 47.2% | Defaultthinking (vendor recommends 'Reasoning effort is set to xhigh' system prompt for reasoning tasks) | Maker's own figureVendor-reportedQwen (Alibaba) ↗ | 16 May 2026 | Claude Code | |
| SciCode | 53.5% | Defaultthinking (vendor recommends 'Reasoning effort is set to xhigh' system prompt for reasoning tasks) | Maker's own figureVendor-reportedQwen (Alibaba) ↗ | 16 May 2026 | — | |
| SimpleBench | 70.4% | DefaultQwen 3.7 Max | Independent testIndependentSimpleBench ↗ | 22 May 2026 | — | AVG@5, temp 0.7; rank 17th; human baseline 83.7%; data https://simple-bench.com/static/js/leaderboard-data.js |
| SimpleQA Verified | 55.8% | Defaultqwen3.7-max | Independent testIndependentEpoch AI ↗ | 27 Aug 2026 | — | Epoch-run (no tools); ±1.57 stderr |
| SkillsBench | 59.2% | Defaultthinking (vendor recommends 'Reasoning effort is set to xhigh' system prompt for reasoning tasks) | Maker's own figureVendor-reportedQwen (Alibaba) ↗ | 16 May 2026 | OpenCode | 78 tasks, avg of 5 runs |
| SWE-bench Multilingual | 78.3% | Defaultthinking (vendor recommends 'Reasoning effort is set to xhigh' system prompt for reasoning tasks) | Maker's own figureVendor-reportedQwen (Alibaba) ↗ | 16 May 2026 | Internal scaffold (bash + file-edit) | |
| SWE-Bench Pro (public, v1) | 60.6% | Defaultthinking (vendor recommends 'Reasoning effort is set to xhigh' system prompt for reasoning tasks) | Maker's own figureVendor-reportedQwen (Alibaba) ↗ | 16 May 2026 | Internal scaffold (bash + file-edit) | Qwen-corrected ('refined') SWE-bench Pro |
| SWE-bench Verified | 77.3% | Defaultqwen3.7-max | Independent testIndependentEpoch AI ↗ | 18 Jun 2026 | — | Epoch-run SWE-bench Verified (Epoch scaffold); no runs after 2026-06; stderr 1.9pt; data https://epoch.ai/data/benchmark_data.zip (swe_bench_verified.csv) |
| SWE-bench Verified | 80.4% | Defaultthinking (vendor recommends 'Reasoning effort is set to xhigh' system prompt for reasoning tasks) | Maker's own figureVendor-reportedQwen (Alibaba) ↗ | 16 May 2026 | Internal scaffold (bash + file-edit) | 200K ctx |
| Terminal-Bench 2.0 | 69.7% | Defaultthinking (vendor recommends 'Reasoning effort is set to xhigh' system prompt for reasoning tasks) | Maker's own figureVendor-reportedQwen (Alibaba) ↗ | 16 May 2026 | Terminus 2 (Harbor) | 5h timeout, max_tokens=80K, 256K ctx, avg of 5 runs; model decides per turn whether to think |
| Terminal-Bench 2.1 | 74.5% | Defaultthinking (setting not stated) | Maker's own figureVendor-reportedQwen (Alibaba) ↗ | 3 Aug 2026 | Claude Code | Qwen3.7-Max column in the Qwen3.8-Max launch table |
| Toolathlon-Verified | 49.7% | Defaultthinking (setting not stated) | Maker's own figureVendor-reportedQwen (Alibaba) ↗ | 3 Aug 2026 | — | Qwen3.7-Max column in the Qwen3.8-Max launch table |
| Vals TaxEval v2 | 75.3% | Defaultvals id alibaba/qwen3.7-max | Independent testIndependentVals.ai ↗ | 1 Sep 2026 | — | vals id alibaba/qwen3.7-max; rank 19/145; ±0.839 stderr; $0.028122/test |