Qwen3.7 Plus vs Qwen3 8
Qwen3.7 Plus Verdict
- Qwen3 8 (max) leads on quality: 66.5 vs 59.9.
- Qwen3.7 Plus is stronger in instruction following, knowledge.
- Qwen3 8 (max) is stronger in agents & tools, coding, composite, human preference, long context, maths, multimodal, reasoning.
- Qwen3.7 Plus is 2.4× cheaper ($1.13 vs $2.67 per 1M blended).
- Qwen3.7 Plus streams 1.7× faster (66 vs 39 tokens per second).
| Metric | Qwen3.7 Plus | Qwen3 8 (max) |
|---|---|---|
| BenchLeader Index | 59.9 | 66.5 |
| Agents & tools score | 45.2 | 68.3 |
| Coding score | 46.7 | 66.5 |
| Composite score | 61.7 | 74.3 |
| Human preference score | 65.2 | 68.2 |
| Instruction following score | 75.6 | – |
| Knowledge score | 64.7 | 63.5 |
| Long context score | 62.8 | 66.6 |
| Maths score | 65.6 | 67.0 |
| Multimodal score | 63.9 | 67.4 |
| Reasoning score | 62.4 | 76.3 |
| Blended price $/M | $1.13 | $2.67 |
| Output speed | 66 tok/s | 39 tok/s |
| Time to first answer | 32.4 s | 55.1 s |
| Context window | 1M | 1M |
| GPQA Diamond | 87.9% | – |
| OTIS Mock AIME | 93.3% | – |
| OSWorld-Verified 2.0 | 2.8% | – |
| SciCode | 45.5% | 52.9% |
| FrontierCode | 10.2% | – |
| ProofBench | – | 58.0% |
| Epoch Capabilities Index | 147.4 | 156.6 |
| LMArena Text | 1456 | 1481 |
| LMArena Hard Prompts | 1473 | 1503 |
| LMArena Coding | 1501 | 1522 |
| LMArena WebDev | – | 1671 |
| LMArena Vision | 1278 | 1315 |
| LMArena Agent | -6.1 | 3.3 |
| LiveBench | – | 78.5% |
| LiveBench Reasoning | – | 88.2% |
| LiveBench Coding | – | 72.9% |
| LiveBench Agentic Coding | – | 64.7% |
| LiveBench Mathematics | – | 91.3% |
| LiveBench Data Analysis | – | 78.4% |
| LiveBench Language | – | 79.7% |
| LiveBench Instruction Following | – | 74.1% |
| AA Intelligence Index | 25.8 | 45.4 |
| IFBench | 78.0% | – |
| AA-LCR | 73.0% | 80.3% |
| MMMU-Pro | 80.5% | 82.8% |
| AA-Omniscience | 1.1 | 12.0 |
| Terminal-Bench Hard | 47.0% | – |
| GPQA Diamond (AA) | 90.0% | 92.8% |
| Humanity's Last Exam (AA) | 35.6% | 43.1% |
| SciCode (AA) | 46.1% | 53.2% |
| τ²-Bench Telecom (AA) | 93.0% | – |
| LiveCodeBench | – | 87.8% |
| MMLU-Pro | – | 88.6% |
| IOI | – | 68.9% |
| LegalBench | – | 83.6% |
| CorpFin | – | 65.8% |
| TaxEval | – | 75.5% |
| Terminal-Bench 2.1 (Vals) | 52.8% | 67.4% |
| SWE-bench (Vals) | – | 85.6% |
| GPQA Diamond (Vals) | – | 93.7% |
| Vals Index | 38.6 | 51.8 |
| CritPt | 9.1% | 20.0% |
| GDPval (AA) | 19.3% | 58.2% |
| τ²-Bench Banking (AA) | 17.5% | 51.3% |
| APEX-Agents (AA) | 22.4% | – |
| Code Migration | 12.9% | 24.0% |
| CyberBench | 37.5% | – |
| Excel Modeling Benchmark | 49.4% | 60.1% |
| Finance Agent v2 | 38.2% | 50.6% |
| Harvey's Legal Agent Benchmark | 0.0% | 10.4% |
| Legal Research Bench | 16.4% | 47.6% |
| MedCode | – | 40.7% |
| MedScribe | – | 85.0% |
| MMMU-Pro (Vals) | – | 88.0% |
| MortgageTax | 66.2% | 64.0% |
| MysteryMechanism | – | 23.9% |
| ProgramBench | – | 0.0% |
| Public Benefits Bench | – | 67.1% |
| SAGE | 39.3% | 51.3% |
| SkillsBench | 54.3% | 42.0% |
| Tax Agent Bench | – | 66.0% |
| Terminal-Bench 4.0 (Vals) | – | 24.8% |
| Terminal-Bench Science | – | 1.4% |
| Vals Multimodal Index | 53.9% | 65.4% |
| Vibe Code Bench 1-100 | – | 12.8% |
| Vibe Code Bench v1.1 | 46.4% | 64.7% |
| LMArena Maths | 1466 | 1498 |
| LMArena Creative Writing | 1432 | 1472 |
| LMArena Instruction Following | 1447 | 1477 |
| LMArena Multi-turn | 1465 | 1495 |
| LMArena Longer Queries | 1465 | 1492 |
| LMArena Document | 1453 | – |
| Chess Puzzles | 24.0% | – |
| Mystery Game Puzzles | 17.0% | – |
| LMCA | 37.6% | – |
| DTBench | 84.0% | – |
Data as of 2026-09-19. Best configuration of each model; every score links to its source on the model pages.
Qwen3.7 Plus vs Qwen3 8: questions
- Is Qwen3.7 Plus better than Qwen3 8?
- Qwen3 8 (max) leads on quality: 66.5 vs 59.9. The BenchLeader Index combines every independent quality benchmark; Qwen3 8 (max) is ahead overall as of 2026-09-19, but check the category scores for your use.
- Is Qwen3.7 Plus better than Qwen3 8 for coding?
- Qwen3 8 scores higher in coding (67 vs 47 on the category index, where 50 is average).
- Is Qwen3.7 Plus better than Qwen3 8 for agentic tasks?
- Qwen3 8 scores higher in agentic tasks (68 vs 45 on the category index, where 50 is average).
- Which is cheaper, Qwen3.7 Plus or Qwen3 8?
- Qwen3.7 Plus is cheaper: $1.13 against $2.67 per million tokens, blended at three input tokens per output token.
- Which is faster, Qwen3.7 Plus or Qwen3 8?
- Qwen3.7 Plus streams faster: 66 against 39 output tokens per second.
- Which has the larger context window?
- Both accept 1M tokens of context.