Across 5 comparable benchmarks, Qwen3.7 Max has better reported results on 5 and worse on 0.
Moderate evidence5 of 5 shared results countedHow we compare
Price basis: Standard rates vs listed rates. The price section below shows the applicable periods.
| Fact | o3-pro | Qwen3.7 Max |
|---|---|---|
| Context window | 200K tokens | 1M tokens |
| Input / 1M | $20.00 | $2.50 |
| Output / 1M | $80.00 | $7.50 |
| Blended (4:1) | $32.00 | $2.06 |
| Status | Deprecated | Available |
| Released | June 10, 2025 | May 20, 2026 |
Blended is 4:1 input to output with 90% cache hits at listed rates — set your own mix.
Mean of 3 shared benchmarks grouped by capability on a 0–100 scale. How averages are built
Qwen3.7 Max ahead on 1 of 1 counted
Qwen3.7 Max ahead on 1 of 1 counted
Qwen3.7 Max ahead on 1 of 1 counted
No shared comparable results for Abstract reasoning, Mathematics, Research & search, Long context, Multimodal, Professional work & productivity, Cybersecurity.
Published scores, with matching modes where available. Gaps show o3-pro relative to Qwen3.7 Max. Reading the results
| Benchmark | o3-pro | Qwen3.7 Max | Gap |
|---|---|---|---|
| SWE-bench VerifiedRepository Engineering · Accuracy | 75.0Thinking High · no tools | 80.4≠Thinking · tools | −5.4 pts#37 vs #14 of 108 |
| GPQA DiamondScientific Reasoning · Accuracy | 84.0Normal · no tools | 92.4≠Max · no tools | −8.4 pts#88 vs #38 of 200 |
| MCP-AtlasTool Orchestration · Pass rate / claim coverage | 44.5Normal · tools | 76.4≠Thinking · tools | −31.9 pts#54 vs #27 of 55 |
| HLEKnowledge Exams · Accuracy | 21.0Normal · no tools | 53.5≠Thinking · tools | −32.5 pts#101 vs #35 of 151 |
| ECICapability Indices · ECI score (capability index, higher is better) | 147Reported best | 154Reported best | 0.96×#59 vs #36 of 168 |
Repository Engineering · Accuracy
Scientific Reasoning · Accuracy
Tool Orchestration · Pass rate / claim coverage
A dash means no result is available.
| Benchmark | o3-pro | Qwen3.7 Max |
|---|---|---|
| AIME 2024 | 93.0 | — |
| AIME2025 | 93.0 | — |
| CritPt | — | 13.4 |
| FrontierMath Tier 4 v2 | — | 34.1 |
| FrontierMath v2 | — | 64.6 |
| IMO-AnswerBench | — | 90.0 |
| MMLU-Pro | — | 89.6 |
| Riemann-bench | — | 15.2 |
| SciCode | — | 53.5 |
| Aider-Polyglot | 84.9 | — |
| Code Migration | — | 13.1 |
| LiveCodeBench | — | 91.6 |
| SWE-bench Multilingual | — | 78.3 |
| SWE-Bench Pro - Public | — | 60.6 |
| Terminal Bench 2.0 | — | 69.7 |
| Terminal Bench Hard | — | 50.8 |
| Terminal-Bench 4.0 | — | 1.5 |
| Vibe Code Bench v1.1 | — | 47.7 |
| WeirdML v2 | 58.2 | — |
| Agents' Last Exam v1.0 | — | 11.8 |
| PinchBench v2 | — | 92.5 |
| LiveBench | — | 73.1 |
| Text Arena (Coding) | — | 1541 |
| IF Bench | — | 80.5 |
| Context Arena | — | 56.0 |
| EBR-bench | — | 9.5 |
| Fiction.liveBench | 97.2 | — |
| GDP.pdf | — | 8.6 |
| AA-AnalystAgent | — | 18.8 |
| EMB (Excel Modeling) | — | 57.0 |
| Finance Agent v2 | — | 47.8 |
| Harvey Lab-AA | — | 83.4 |
| Legal Research Bench | — | 25.5 |
| Tax Agent Bench | — | 15.8 |
| τ³-Banking | — | 11.8 |
| ARC-AGI-1 | 59.3 | — |
| ARC-AGI-2 | 4.9 | — |
| SimpleBench | — | 70.4 |
| MedCode | — | 38.8 |
| MedScribe | — | 79.4 |
Time-of-day providers use peak-hour rates; fixed-price providers use standard rates. Off-peak discounts remain in the published details.
Drag the handle above the bar for cache hits, the one below it for the input : output mix · 5% steps
Blended price = effective input rate at 90% cache hits × 80% + output rate × 20%
o3-pro costs1,453%morethan Qwen3.7 Max per million tokens — 15.5× the price.Using standard rates vs listed rates at this token mix.
o3-pro: Standard pricing. OpenAI official pricing (developers.openai.com/api/docs/pricing) verified 2026-11-13; standard tier; Deprecated; shutdown 2026-12-11 (replacement: gpt-6.1-sol pro)
Qwen3.7 Max: International pricing. context_length <= 1000000 Alibaba Model Studio International (Singapore)
| Price / 1M tokens | o3-pro | Qwen3.7 Max | Ratio |
|---|---|---|---|
| Input | $20.00 | $2.50 | 8.00× |
| Output | $80.00 | $7.50 | 10.7× |
| Cached input | — | $0.50 | — |
| Cache write | — | — | — |
| Blended | $32.00 | $2.06 | 15.5× |
Estimates use the selected token mix. Cache writes, tool fees and long-context surcharges are excluded.
Reported cost per task, paired with the score from the same run. Settings and sources may differ; costs do not affect the verdict.
| Benchmark | o3-pro | Qwen3.7 Max |
|---|---|---|
| Agents' Last Exam v1.0 | Not reported | |
| ARC-AGI-1 | 1 more settings | Not reported |
| ARC-AGI-2 | $7.5516 / task Average · ARC-AGI-2 Score: 4.9 Thinking High (With Tools) $4.7441 / task Average · ARC-AGI-2 Score: 1.9 Thinking Medium (With Tools) | Not reported |
| Code Migration | Not reported | $1.9329 / task Average · Hidden-test pass rate (%): 13.1 Unspecified Reasoning (With Tools) |
| EMB (Excel Modeling) | Not reported | |
| Finance Agent v2 | Not reported | |
| Legal Research Bench | Not reported | |
| MedCode | Not reported | |
| MedScribe | Not reported | |
| Tax Agent Bench | Not reported | $0.4355 / task Average · Accuracy (%, rubric × citation quality): 15.8 Unspecified Reasoning (With Tools) |
| Vibe Code Bench v1.1 | Not reported |
| Attribute | o3-pro | Qwen3.7 Max |
|---|---|---|
| Availability | Deprecated | Available |
| Provider | OpenAI | Alibaba |
| Model type | Reasoning model | Reasoning model |
| First release | June 10, 2025 | May 20, 2026 |
| Context window | 200K tokens | 1M tokens |
| Max output | 100K tokens | 64K tokens |
| Architecture | Undisclosed | Undisclosed |
| Documented capabilities | Reasoning | Reasoning, Multilingual, Coding, Function calling, Tool use, Long context |
| Knowledge cutoff | 2024-06-01 | — |
| Input modalities | Text, Image | Text |
| Output modalities | Text | Text |
| Weights | Closed / API only | Closed / API only |
| Commercial use | Proprietary | Proprietary |
These are reported-score comparisons, not controlled head-to-head tests.
What people found comparing these models, or using either one. Comments also appear on the models' own pages.
Used o3-pro or Qwen3.7 Max? Sign in with Google to share what worked, what did not, and which API or setup you tested.
No community comments on either model yet.
Alternatives with the most benchmark overlap for this matchup.