Model comparison
Claude 3.5 Sonnet vs GPT-5.6 Luna
GPT-5.6 Luna is the stronger model overall, scoring 54.6 to 34.6 on the Noometry Index.
Last verified . 26 shared benchmarks.
Summary
- They share 26 benchmarks with published results for both. Claude 3.5 Sonnet scores higher in 0 categories and GPT-5.6 Luna in 10 categories; 10 gaps are clear of the uncertainty.
- The widest gap is in math, where GPT-5.6 Luna leads 77.7 to 19.2.
- The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 8.5% for Claude 3.5 Sonnet and 98.3% for GPT-5.6 Luna.
Side by side
| Claude 3.5 Sonnet | GPT-5.6 Luna | |
|---|---|---|
| Provider | Anthropic | OpenAI |
| Noometry Index | 34.6 | 54.6 |
| Released | 2024-06-20 | 2026-07-09 |
| Weights | Proprietary | Proprietary |
| Context window | — | 1.05M |
| Max output | — | 128K |
| Input $ / M tokens | — | $0.20 |
| Output $ / M tokens | — | $1.20 |
| Results tracked | 60 | 52 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding GPT-5.6 Luna leads
Claude 3.5 Sonnet: 39.0 (#165), GPT-5.6 Luna: 54.5 (#28)
| Benchmark | Claude 3.5 Sonnet | GPT-5.6 Luna |
|---|---|---|
| WeirdML | 40% | 60.9% |
| LMArena Coding | 1342 | 1466 |
| DeepSWE | — | 67.2% |
| FrontierCode | — | 39.8% |
| Aider Polyglot | 51.6% | — |
| CursorBench | — | 35.9% |
| LMArena WebDev | — | 1519 |
| SciCode | — | 53.6% |
| GSO | 4.6% | — |
| BigCodeBench Instruct | 46.8% | — |
| LiveBench Coding | 67.1% | — |
| BigCodeBench Complete | 58.6% | — |
| CadEval | 48% | — |
| ALE-Bench | — | 1,667 |
| HumanEval+ | 81.7% | — |
| MBPP+ | 74.3% | — |
Agentic & Tool Use GPT-5.6 Luna leads
Claude 3.5 Sonnet: 32.3 (#67), GPT-5.6 Luna: 34.4 (#45)
| Benchmark | Claude 3.5 Sonnet | GPT-5.6 Luna |
|---|---|---|
| BALROG | 32.6% | 45.6% |
| APEX-Agents | — | 43% |
| TheAgentCompany | 24% | — |
| Cybench | 17.5% | — |
| GDP.pdf | — | 22.7% |
| METR Time Horizons | 45.2% | — |
| Vending-Bench 2 | — | 4,095 |
Reasoning GPT-5.6 Luna leads
Claude 3.5 Sonnet: 23.1 (#183), GPT-5.6 Luna: 47.6 (#43)
| Benchmark | Claude 3.5 Sonnet | GPT-5.6 Luna |
|---|---|---|
| SimpleBench | 41.4% | 46.8% |
| LMArena Hard Prompts | 1305 | 1451 |
| DTBench | 67.8% | 89.1% |
| Epoch Capabilities Index | 133.55 | 156.39 |
| ARC-AGI-2 | — | 59.5% |
| Kagi LLM Benchmark | — | 49.1% |
| NYT Connections (extended) | — | 69.4% |
| ARC-AGI-1 | — | 88% |
| CritPt | — | 20.6% |
| Chess Puzzles | — | 40% |
| EnigmaEval | 0.9% | — |
| LiveBench Reasoning | 56.7% | — |
| Mystery Game Puzzles | — | 21% |
| LiveBench Data Analysis | 55% | — |
| LMCA | — | 48.5% |
| Surface Evolver Bench | — | 61.9% |
| ForecastBench | 60.7 | — |
| LiveBench | 59% | — |
Math GPT-5.6 Luna leads
Claude 3.5 Sonnet: 19.2 (#288), GPT-5.6 Luna: 77.7 (#14)
| Benchmark | Claude 3.5 Sonnet | GPT-5.6 Luna |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 8.5% | 98.3% |
| LMArena Math | 1307 | 1458 |
| FrontierMath (Tiers 1-3) | — | 82.1% |
| FrontierMath Tier 4 | — | 61% |
| ProofBench | — | 60% |
| Omni-MATH | 27.6% | — |
| LiveBench Math | 52.3% | — |
| MATH Level 5 | 56.9% | — |
| FrontierMath (Feb 2025 set) | 2.1% | — |
| FrontierMath Tier 4 (v1) | 0% | — |
Knowledge GPT-5.6 Luna leads
Claude 3.5 Sonnet: 28.6 (#245), GPT-5.6 Luna: 58.5 (#34)
| Benchmark | Claude 3.5 Sonnet | GPT-5.6 Luna |
|---|---|---|
| GPQA Diamond | 55.3% | 91.6% |
| LMArena Expert | 1265 | 1478 |
| Humanity's Last Exam | 4.1% | — |
| SimpleQA Verified | — | 41% |
| MMLU-Pro | 77.7% | — |
| Confabulations | 19.9% | — |
| GPQA (HELM) | 56.5% | — |
| MMLU | 87.3% | — |
Multimodal GPT-5.6 Luna leads
Claude 3.5 Sonnet: 26.5 (#120), GPT-5.6 Luna: 42.7 (#28)
| Benchmark | Claude 3.5 Sonnet | GPT-5.6 Luna |
|---|---|---|
| LMArena Vision | 1125 | 1258 |
| Video-MME | 60% | — |
| GeoBench | 62% | — |
| VPCT | 33% | — |
| Blueprint-Bench 2 | — | 22.6% |
| Furniture Assembly | — | 42.5% |
| LMArena Document | — | 1457 |
Multilingual GPT-5.6 Luna leads
Claude 3.5 Sonnet: 43.2 (#185), GPT-5.6 Luna: 52.8 (#78)
| Benchmark | Claude 3.5 Sonnet | GPT-5.6 Luna |
|---|---|---|
| LMArena Non-English | 1283 | 1417 |
| LMArena Chinese | 1272 | 1470 |
| LMArena French | 1305 | 1456 |
| LMArena German | 1297 | 1454 |
| LMArena Japanese | 1234 | 1411 |
| LMArena Korean | 1200 | 1415 |
| LMArena Russian | 1306 | 1428 |
| LMArena Spanish | 1290 | 1448 |
Instruction Following GPT-5.6 Luna leads
Claude 3.5 Sonnet: 68.8 (#182), GPT-5.6 Luna: 75.6 (#57)
| Benchmark | Claude 3.5 Sonnet | GPT-5.6 Luna |
|---|---|---|
| LMArena Instruction Following | 1297 | 1437 |
| LiveBench Instruction Following | 69.3% | — |
| IFEval | 85.5% | — |
Long Context GPT-5.6 Luna leads
Claude 3.5 Sonnet: 39.9 (#167), GPT-5.6 Luna: 43.9 (#82)
| Benchmark | Claude 3.5 Sonnet | GPT-5.6 Luna |
|---|---|---|
| LMArena Longer Query | 1311 | 1436 |
Writing & Preference GPT-5.6 Luna leads
Claude 3.5 Sonnet: 52.9 (#164), GPT-5.6 Luna: 68.0 (#29)
| Benchmark | Claude 3.5 Sonnet | GPT-5.6 Luna |
|---|---|---|
| LMArena Text | 1298 | 1431 |
| LMArena Creative Writing | 1292 | 1396 |
| EQ-Bench Creative Writing | 1451 | 1829 |
| LMArena Multi-Turn | 1326 | 1434 |
| Short-Story Creative Writing | 80.3% | — |
| WildBench | 79.2% | — |
| EQ-Bench 4 | — | 1156 |
| LiveBench Language | 53.8% | — |
Frequently asked questions
Is Claude 3.5 Sonnet better than GPT-5.6 Luna?
GPT-5.6 Luna is the stronger model overall, scoring 54.6 to 34.6 on the Noometry Index.
Is Claude 3.5 Sonnet or GPT-5.6 Luna better for coding?
GPT-5.6 Luna scores higher on coding benchmarks: 54.5 versus 39.0 in the Noometry coding category.
How many benchmarks do Claude 3.5 Sonnet and GPT-5.6 Luna share?
26 benchmarks have published results for both models. Claude 3.5 Sonnet has 60 scored results on Noometry and GPT-5.6 Luna has 52.