GPT-6 Sol vs Qwen3.8 Max (0902)
GPT-6 Sol 
Qwen3.8 Max (0902) Verdict
- GPT-6 Sol (max) leads on quality: 67.7 vs 66.0.
- GPT-6 Sol (max) is stronger in composite, knowledge, long context, multimodal, reasoning.
- Qwen3.8 Max (0902) (max) is stronger in agents & tools, coding, human preference, maths.
- Qwen3.8 Max (0902) (max) is 1.3× cheaper ($3.00 vs $4.00 per 1M blended).
- GPT-6 Sol (max) streams 3.2× faster (126 vs 39 tokens per second).
| Metric | GPT-6 Sol (max) | Qwen3.8 Max (0902) (max) |
|---|---|---|
| BenchLeader Index | 67.7 | 66.0 |
| Composite score | 86.2 | 72.6 |
| Knowledge score | 75.4 | 63.1 |
| Long context score | 67.7 | 66.0 |
| Multimodal score | 67.1 | 67.0 |
| Reasoning score | 95.0 | 74.1 |
| Agents & tools score | – | 68.3 |
| Coding score | – | 66.4 |
| Human preference score | – | 68.3 |
| Maths score | – | 67.1 |
| Blended price $/M | $4.00 | $3.00 |
| Output speed | 126 tok/s | 39 tok/s |
| Time to first answer | 107.2 s | 54.8 s |
| Context window | 1.1M | 1M |
| SciCode | – | 52.9% |
| ProofBench | – | 58.0% |
| Epoch Capabilities Index | – | 156.7 |
| LMArena Text | – | 1481 |
| LMArena Hard Prompts | – | 1503 |
| LMArena Coding | – | 1522 |
| LMArena WebDev | – | 1671 |
| LMArena Vision | – | 1315 |
| LMArena Agent | – | 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 | 47.5 | 45.4 |
| AA-LCR | 83.7% | 80.3% |
| MMMU-Pro | 83.3% | 82.8% |
| AA-Omniscience | 27.1 | 12.0 |
| GPQA Diamond (AA) | – | 92.8% |
| Humanity's Last Exam (AA) | 47.9% | 43.1% |
| SciCode (AA) | 57.6% | 53.2% |
| LiveCodeBench | – | 87.8% |
| MMLU-Pro | – | 88.6% |
| IOI | – | 68.9% |
| LegalBench | – | 83.6% |
| CorpFin | – | 65.8% |
| TaxEval | – | 75.5% |
| Terminal-Bench 2.1 (Vals) | – | 67.4% |
| SWE-bench (Vals) | – | 85.6% |
| GPQA Diamond (Vals) | – | 93.7% |
| Vals Index | – | 51.8 |
| CritPt | 30.9% | 20.0% |
| GDPval (AA) | 49.4% | 58.4% |
| τ²-Bench Banking (AA) | – | 51.3% |
| Code Migration | – | 24.0% |
| Excel Modeling Benchmark | – | 60.1% |
| Finance Agent v2 | – | 50.6% |
| Harvey's Legal Agent Benchmark | – | 10.4% |
| Legal Research Bench | – | 47.6% |
| MedCode | – | 40.7% |
| MedScribe | – | 85.0% |
| MMMU-Pro (Vals) | – | 88.0% |
| MortgageTax | – | 64.0% |
| MysteryMechanism | – | 23.9% |
| ProgramBench | – | 0.0% |
| Public Benefits Bench | – | 67.1% |
| SAGE | – | 51.3% |
| SkillsBench | – | 42.0% |
| Tax Agent Bench | – | 66.0% |
| Terminal-Bench 4.0 (Vals) | – | 24.8% |
| Terminal-Bench Science | – | 1.4% |
| Vals Multimodal Index | – | 65.4% |
| Vibe Code Bench 1-100 | – | 12.8% |
| Vibe Code Bench v1.1 | – | 64.7% |
| LMArena Maths | – | 1498 |
| LMArena Creative Writing | – | 1472 |
| LMArena Instruction Following | – | 1477 |
| LMArena Multi-turn | – | 1495 |
| LMArena Longer Queries | – | 1492 |
| Terminal-Bench 4.0 (AA) | 43.9% | 38.9% |
| Terminal-Bench 2.1 (AA) | – | 88.8% |
| AutomationBench | 61.6% | 56.2% |
| GDP.pdf | 24.8% | 22.8% |
| MLCR | 16.1% | 20.0% |
| Harvey LAB | – | 93.7% |
| AA-Omniscience: accuracy | 54.5% | 31.9% |
| AA-Omniscience: non-hallucination | 39.9% | 71.2% |
| AA-Briefcase | 1483 | 1640 |
Data as of 2026-09-23. Best configuration of each model; every score links to its source on the model pages.
GPT-6 Sol vs Qwen3.8 Max (0902): questions
- Is GPT-6 Sol better than Qwen3.8 Max (0902)?
- GPT-6 Sol (max) leads on quality: 67.7 vs 66.0. The BenchLeader Index combines every independent quality benchmark; GPT-6 Sol (max) is ahead overall as of 2026-09-23, but check the category scores for your use.
- Which is cheaper, GPT-6 Sol or Qwen3.8 Max (0902)?
- Qwen3.8 Max (0902) is cheaper: $3.00 against $4.00 per million tokens, blended at three input tokens per output token.
- Which is faster, GPT-6 Sol or Qwen3.8 Max (0902)?
- GPT-6 Sol streams faster: 126 against 39 output tokens per second.
- Which has the larger context window?
- GPT-6 Sol accepts more context: 1.1M against 1M tokens.