GPT-5.5 vs GPT-6 Luna
GPT-5.5 Verdict
- GPT-5.5 (xhigh) leads on quality: 66.9 vs 61.2.
- GPT-5.5 (xhigh) is stronger in agents & tools, coding, instruction following, knowledge, long context, maths, multimodal, reasoning.
- GPT-6 Luna (max) is stronger in composite.
- GPT-6 Luna (max) is 56× cheaper ($0.200 vs $11.25 per 1M blended).
- GPT-6 Luna (max) streams 1.5× faster (157 vs 102 tokens per second).
| Metric | GPT-5.5 (xhigh) | GPT-6 Luna (max) |
|---|---|---|
| BenchLeader Index | 66.9 | 61.2 |
| Agents & tools score | 61.9 | – |
| Coding score | 64.3 | – |
| Composite score | 71.0 | 73.9 |
| Instruction following score | 73.8 | – |
| Knowledge score | 66.0 | 63.2 |
| Long context score | 68.0 | 67.5 |
| Maths score | 71.1 | – |
| Multimodal score | 63.6 | 59.2 |
| Reasoning score | 78.1 | 76.1 |
| Blended price $/M | $11.25 | $0.200 |
| Output speed | 102 tok/s | 157 tok/s |
| Time to first answer | 64.1 s | 142.6 s |
| Context window | 1.1M | 1.1M |
| FrontierMath Tiers 1–3 | 85.3% | – |
| FrontierMath Tier 4 | 72.5% | – |
| SimpleQA Verified | 63.0% | – |
| OSWorld-Verified 2.0 | 13.0% | – |
| SciCode | 56.1% | – |
| WeirdML | 84.9% | – |
| ProofBench | 50.0% | – |
| GSO-Bench | 40.2% | – |
| LMArena WebDev | 1510 | – |
| LMArena Agent | 5 | – |
| LiveBench | 80.2% | – |
| LiveBench Reasoning | 89.7% | – |
| LiveBench Coding | 82.2% | – |
| LiveBench Agentic Coding | 54.0% | – |
| LiveBench Mathematics | 95.9% | – |
| LiveBench Data Analysis | 81.6% | – |
| LiveBench Language | 87.4% | – |
| LiveBench Instruction Following | 70.7% | – |
| AA Intelligence Index | 38.4 | 37.3 |
| IFBench | 75.8% | – |
| AA-LCR | 84.3% | 83.3% |
| MMMU-Pro | 79.9% | 75.5% |
| AA-Omniscience | 20.5 | 0.7 |
| Terminal-Bench Hard | 60.6% | – |
| GPQA Diamond (AA) | 93.5% | – |
| Humanity's Last Exam (AA) | 45.8% | 38.5% |
| SciCode (AA) | 55.8% | 54.6% |
| τ²-Bench Telecom (AA) | 93.9% | – |
| LiveCodeBench | 85.3% | – |
| MMLU-Pro | 88.1% | – |
| LegalBench | 86.5% | – |
| CorpFin | 68.4% | – |
| TaxEval | 75.0% | – |
| SWE-bench (Vals) | 82.6% | – |
| GPQA Diamond (Vals) | 93.2% | – |
| Vals Index | 57.4 | – |
| AIME 2026 | 100.0% | – |
| HMMT February 2026 | 98.5% | – |
| MathArena Apex | 80.2% | – |
| MCP Atlas | 75.3% | – |
| ARC-AGI-1 | 95.0% | – |
| ARC-AGI-2 | 85.0% | – |
| CritPt | 27.1% | 19.4% |
| GDPval (AA) | 41.8% | 43.4% |
| τ²-Bench Banking (AA) | 39.0% | – |
| ITBench SRE (AA) | 45.8% | – |
| Analyst Agent (AA) | 50.0% | – |
| APEX-Agents (AA) | 37.7% | – |
| CaseLaw v2 | 66.2% | – |
| Code Migration | 45.2% | – |
| Excel Modeling Benchmark | 64.5% | – |
| Finance Agent v2 | 51.8% | – |
| Harvey's Legal Agent Benchmark | 3.8% | – |
| Legal Research Bench | 40.4% | – |
| MedCode | 49.1% | – |
| MedScribe | 86.9% | – |
| MMMU-Pro (Vals) | 88.3% | – |
| MortgageTax | 68.8% | – |
| ProgramBench | 0.5% | – |
| Public Benefits Bench | 60.9% | – |
| SAGE | 51.5% | – |
| SkillsBench | 62.2% | – |
| SREBench | 3.8% | – |
| Tax Agent Bench | 60.5% | – |
| Time Horizon Index: KSP | 8.3% | – |
| Vals Multimodal Index | 68.1% | – |
| Vibe Code Bench v1.1 | 69.8% | – |
| DrugDiscoveryBench | 51.6% | – |
| FORTRESS | 16.3% | – |
| SWE Atlas: Codebase QnA | 45.4% | – |
| SWE Atlas: Refactoring | 44.8% | – |
| SWE Atlas: Test Writing | 42.6% | – |
| EBR-bench | 34.3% | – |
| Mystery Game Puzzles | 56.0% | – |
| PostTrainBench | 27.2% | – |
| DeepSWE | 67.0% | – |
| LMCA | 54.3% | – |
| DTBench | 96.0% | – |
| ALE-Bench | 1943.0 | – |
| GDP.pdf | 26.0% | – |
| Terminal-Bench 4.0 (AA) | 14.7% | 12.6% |
| Terminal-Bench 2.1 (AA) | 84.3% | – |
| AutomationBench | – | 53.2% |
| GDP.pdf | – | 20.4% |
| MLCR | – | 16.1% |
| AA-Omniscience: accuracy | 58.0% | 43.8% |
| AA-Omniscience: non-hallucination | 11.0% | 23.3% |
| AA-Briefcase | – | 1299 |
Data as of 2026-09-23. Best configuration of each model; every score links to its source on the model pages.
GPT-5.5 vs GPT-6 Luna: questions
- Is GPT-5.5 better than GPT-6 Luna?
- GPT-5.5 (xhigh) leads on quality: 66.9 vs 61.2. The BenchLeader Index combines every independent quality benchmark; GPT-5.5 (xhigh) is ahead overall as of 2026-09-23, but check the category scores for your use.
- Which is cheaper, GPT-5.5 or GPT-6 Luna?
- GPT-6 Luna is cheaper: $0.200 against $11.25 per million tokens, blended at three input tokens per output token.
- Which is faster, GPT-5.5 or GPT-6 Luna?
- GPT-6 Luna streams faster: 157 against 102 output tokens per second.
- Which has the larger context window?
- Both accept 1.1M tokens of context.