GPT-5.5 vs GPT-6.1 Sol
GPT-5.5 Verdict
- GPT-6.1 Sol (max) leads on quality: 69.3 vs 65.9.
- GPT-5.5 (xhigh) is stronger in instruction following, long context.
- GPT-6.1 Sol (max) is stronger in agents & tools, coding, composite, knowledge, maths, multimodal, reasoning, human preference.
- GPT-6.1 Sol (max) is 2.8× cheaper ($4.00 vs $11.25 per 1M blended).
- GPT-5.5 (xhigh) streams 1.6× faster (88 vs 56 tokens per second).
| Metric | GPT-5.5 (xhigh) | GPT-6.1 Sol (max) |
|---|---|---|
| BenchLeader Index | 65.9 | 69.3 |
| Agents & tools score | 61.5 | 63.9 |
| Coding score | 61.4 | 72.4 |
| Composite score | 69.2 | 79.0 |
| Instruction following score | 73.8 | – |
| Knowledge score | 65.5 | 78.8 |
| Long context score | 67.5 | 66.8 |
| Maths score | 69.9 | 72.5 |
| Multimodal score | 63.1 | 66.2 |
| Reasoning score | 75.8 | 76.4 |
| Human preference score | – | 68.1 |
| Blended price $/M | $11.25 | $4.00 |
| Output speed | 88 tok/s | 56 tok/s |
| Time to first answer | 39.0 s | 326.9 s |
| Context window | 1.1M | 1.1M |
| GPQA Diamond | – | 95.4% |
| FrontierMath Tiers 1–3 | 85.3% | 93.7% |
| FrontierMath Tier 4 | 72.5% | 100.0% |
| OTIS Mock AIME | – | 100.0% |
| SimpleQA Verified | 63.0% | 73.9% |
| Terminal-Bench | – | 58.2% |
| OSWorld-Verified 2.0 | 13.0% | – |
| SciCode | 56.1% | 54.2% |
| WeirdML | 84.9% | – |
| APEX-Agents | – | 60.0% |
| ProofBench | 50.0% | – |
| GSO-Bench | 40.2% | – |
| LMArena Text | – | 1484 |
| LMArena Hard Prompts | – | 1507 |
| LMArena Coding | – | 1545 |
| LMArena WebDev | 1513 | 1755 |
| LMArena Vision | – | 1288 |
| LMArena Agent | 4.4 | 11.7 |
| LiveBench | 80.2% | 81.6% |
| LiveBench Reasoning | 89.7% | 92.6% |
| LiveBench Coding | 82.2% | 80.4% |
| LiveBench Agentic Coding | 54.0% | 54.5% |
| LiveBench Mathematics | 95.9% | 96.8% |
| LiveBench Data Analysis | 81.6% | 82.7% |
| LiveBench Language | 87.4% | 90.1% |
| LiveBench Instruction Following | 70.7% | 74.2% |
| AA Intelligence Index v4.3.2 | 38.4 | 51.8 |
| IFBench | 75.8% | – |
| AA-LCR | 84.3% | 83.0% |
| MMMU-Pro | 79.9% | 86.0% |
| AA-Omniscience | 20.5 | 41.5 |
| Terminal-Bench Hard | 60.6% | – |
| GPQA Diamond (AA) | 93.5% | – |
| Humanity's Last Exam (AA) | 45.8% | 52.9% |
| SciCode (AA) | 55.8% | 54.2% |
| τ²-Bench Telecom (AA) | 93.9% | – |
| LiveCodeBench | 85.3% | – |
| MMLU-Pro | 88.1% | – |
| IOI | – | 96.9% |
| LegalBench | 86.5% | – |
| CorpFin | 68.4% | – |
| TaxEval | 75.0% | – |
| SWE-bench (Vals) | 82.6% | – |
| GPQA Diamond (Vals) | 93.2% | – |
| Vals Index | – | 61.1 |
| AIME 2026 | 100.0% | – |
| HMMT February 2026 | 98.5% | – |
| MathArena Apex | 80.2% | – |
| MCP Atlas | 75.3% | – |
| ARC-AGI-1 | 95.0% | 96.5% |
| ARC-AGI-2 | 85.0% | 94.2% |
| ARC-AGI-3 | – | 96.2% |
| CritPt | 27.1% | 31.7% |
| GDPval-AA v2.1 | 42.7% | 53.8% |
| τ³-Banking (AA) | 39.0% | – |
| ITBench SRE (AA) | 45.8% | – |
| Analyst Agent (AA) | 50.0% | 50.0% |
| APEX-Agents (AA) | 37.7% | – |
| BioMysteryBench | – | 79.6% |
| CaseLaw v2 | 66.2% | – |
| Code Migration | 45.2% | 65.1% |
| CyberBench | – | 39.3% |
| Excel Modeling Benchmark | 64.5% | 70.8% |
| Finance Agent v2 | 51.8% | 52.0% |
| Harvey's Legal Agent Benchmark | 3.8% | 5.4% |
| Legal Research Bench | 40.4% | 38.5% |
| MedCode | 49.1% | 48.8% |
| MedScribe | 86.9% | 86.5% |
| MMMU-Pro (Vals) | 88.3% | – |
| MortgageTax | 68.8% | – |
| MysteryMechanism | – | 46.4% |
| ProgramBench | 0.5% | – |
| Public Benefits Bench | 60.9% | 59.3% |
| SAGE | 51.5% | 46.5% |
| SkillsBench | 62.2% | – |
| SREBench | 3.8% | 50.8% |
| Tax Agent Bench | 60.5% | 62.3% |
| Terminal-Bench 4.0 (Vals) | – | 55.0% |
| Time Horizon Index: KSP | 8.3% | – |
| Vals Multimodal Index | 68.1% | – |
| Vibe Code Bench v1.1 | 69.8% | 88.9% |
| FORTRESS | 16.3% | – |
| SWE Atlas: Codebase QnA | 45.4% | – |
| SWE Atlas: Refactoring | 44.8% | – |
| SWE Atlas: Test Writing | 42.6% | – |
| LMArena Maths | – | 1488 |
| LMArena Creative Writing | – | 1462 |
| LMArena Instruction Following | – | 1488 |
| LMArena Multi-turn | – | 1487 |
| LMArena Longer Queries | – | 1496 |
| Chess Puzzles | – | 61.0% |
| EBR-bench | 34.3% | 54.3% |
| Mystery Game Puzzles | 56.0% | 80.0% |
| PostTrainBench | 27.2% | – |
| DeepSWE v1.1 | 67.0% | – |
| LMCA | 54.3% | – |
| DTBench | 96.0% | – |
| ALE-Bench | 1943.0 | – |
| GDP.pdf | 26.0% | – |
| Terminal-Bench 4.0 (AA) | 14.7% | 56.1% |
| Terminal-Bench 2.1 (AA) | 84.3% | – |
| AutomationBench | – | 64.9% |
| GDP.pdf | – | 31.0% |
| MLCR | – | 33.9% |
| Harvey LAB | – | 6.9% |
| AA-Omniscience: accuracy | 58.0% | 62.1% |
| AA-Omniscience: non-hallucination | 11.0% | 45.7% |
| AA-Briefcase v1.1 | – | 1557 |
Data as of 2026-10-11. Best configuration of each model; every score links to its source on the model pages.
GPT-5.5 vs GPT-6.1 Sol: questions
- Is GPT-5.5 better than GPT-6.1 Sol?
- GPT-6.1 Sol (max) leads on quality: 69.3 vs 65.9. The BenchLeader Index combines every independent quality benchmark; GPT-6.1 Sol (max) is ahead overall as of 2026-10-11, but check the category scores for your use.
- Is GPT-5.5 better than GPT-6.1 Sol for coding?
- GPT-6.1 Sol scores higher in coding (72 vs 61 on the category index, where 50 is average).
- Is GPT-5.5 better than GPT-6.1 Sol for agentic tasks?
- GPT-6.1 Sol scores higher in agentic tasks (64 vs 62 on the category index, where 50 is average).
- Which is cheaper, GPT-5.5 or GPT-6.1 Sol?
- GPT-6.1 Sol is cheaper: $4.00 against $11.25 per million tokens, blended at three input tokens per output token.
- Which is faster, GPT-5.5 or GPT-6.1 Sol?
- GPT-5.5 streams faster: 88 against 56 output tokens per second.
- Which has the larger context window?
- Both accept 1.1M tokens of context.