Claude Opus 5.5 vs Qwen3.7 Plus
Claude Opus 5.5 
Qwen3.7 Plus Verdict
- Claude Opus 5.5 (thinking) leads on quality: 70.7 vs 59.5.
- Claude Opus 5.5 (thinking) is stronger in composite, knowledge, long context, multimodal, reasoning.
- Qwen3.7 Plus is stronger in agents & tools, coding, human preference, instruction following, maths.
- Qwen3.7 Plus is 7.1× cheaper ($1.13 vs $8.00 per 1M blended).
- Qwen3.7 Plus streams 1.4× faster (65 vs 48 tokens per second).
| Metric | Claude Opus 5.5 (thinking) | Qwen3.7 Plus |
|---|---|---|
| BenchLeader Index | 70.7 | 59.5 |
| Composite score | 95.0 | 59.9 |
| Knowledge score | 85.0 | 63.9 |
| Long context score | 68.5 | 62.5 |
| Multimodal score | 71.9 | 63.6 |
| Reasoning score | 95.0 | 62.0 |
| Agents & tools score | – | 45.2 |
| Coding score | – | 46.2 |
| Human preference score | – | 65.2 |
| Instruction following score | – | 75.6 |
| Maths score | – | 65.7 |
| Blended price $/M | $8.00 | $1.13 |
| Output speed | 48 tok/s | 65 tok/s |
| Time to first answer | 4.2 s | 33.0 s |
| Context window | 1M | 1M |
| GPQA Diamond | – | 87.9% |
| OTIS Mock AIME | – | 93.3% |
| OSWorld-Verified 2.0 | – | 2.8% |
| SciCode | – | 45.5% |
| FrontierCode | – | 10.2% |
| Epoch Capabilities Index | – | 147.4 |
| LMArena Text | – | 1456 |
| LMArena Hard Prompts | – | 1473 |
| LMArena Coding | – | 1501 |
| LMArena Vision | – | 1278 |
| LMArena Agent | – | -6.1 |
| AA Intelligence Index | 57.6 | 25.2 |
| IFBench | – | 78.0% |
| AA-LCR | 84.7% | 73.0% |
| MMMU-Pro | 87.7% | 80.5% |
| AA-Omniscience | 46.4 | 1.1 |
| Terminal-Bench Hard | – | 47.0% |
| GPQA Diamond (AA) | – | 90.0% |
| Humanity's Last Exam (AA) | 61.4% | 35.6% |
| SciCode (AA) | 66.9% | 46.1% |
| τ²-Bench Telecom (AA) | – | 93.0% |
| Terminal-Bench 2.1 (Vals) | – | 52.8% |
| Vals Index | – | 38.6 |
| CritPt | 31.7% | 9.1% |
| GDPval (AA) | 67.3% | 12.8% |
| τ²-Bench Banking (AA) | – | 17.5% |
| APEX-Agents (AA) | – | 22.4% |
| Code Migration | – | 12.9% |
| Excel Modeling Benchmark | – | 49.4% |
| Finance Agent v2 | – | 38.2% |
| Harvey's Legal Agent Benchmark | – | 0.0% |
| Legal Research Bench | – | 16.4% |
| MortgageTax | – | 66.2% |
| SAGE | – | 39.3% |
| SkillsBench | – | 54.3% |
| Vals Multimodal Index | – | 53.9% |
| Vibe Code Bench v1.1 | – | 46.4% |
| LMArena Maths | – | 1466 |
| LMArena Creative Writing | – | 1432 |
| LMArena Instruction Following | – | 1447 |
| LMArena Multi-turn | – | 1465 |
| LMArena Longer Queries | – | 1465 |
| LMArena Document | – | 1453 |
| Chess Puzzles | – | 24.0% |
| Mystery Game Puzzles | – | 17.0% |
| LMCA | – | 37.6% |
| DTBench | – | 84.0% |
| Terminal-Bench 4.0 (AA) | 59.6% | 1.0% |
| Terminal-Bench 2.1 (AA) | – | 61.0% |
| AutomationBench | 69.5% | – |
| GDP.pdf | 26.2% | – |
| Harvey LAB | 91.2% | – |
| AA-Omniscience: accuracy | 66.2% | 22.5% |
| AA-Omniscience: non-hallucination | 41.4% | 72.3% |
| AA-Briefcase | 1822 | – |
Data as of 2026-09-22. Best configuration of each model; every score links to its source on the model pages.
Claude Opus 5.5 vs Qwen3.7 Plus: questions
- Is Claude Opus 5.5 better than Qwen3.7 Plus?
- Claude Opus 5.5 (thinking) leads on quality: 70.7 vs 59.5. The BenchLeader Index combines every independent quality benchmark; Claude Opus 5.5 (thinking) is ahead overall as of 2026-09-22, but check the category scores for your use.
- Which is cheaper, Claude Opus 5.5 or Qwen3.7 Plus?
- Qwen3.7 Plus is cheaper: $1.13 against $8.00 per million tokens, blended at three input tokens per output token.
- Which is faster, Claude Opus 5.5 or Qwen3.7 Plus?
- Qwen3.7 Plus streams faster: 65 against 48 output tokens per second.
- Which has the larger context window?
- Both accept 1M tokens of context.