Model comparison
GPT-4o vs GPT-5.6 Luna
GPT-5.6 Luna is the stronger model overall, scoring 54.7 to 28.5 on the Noometry Index.
Last verified . 32 shared benchmarks.
Summary
- They share 32 benchmarks with published results for both. GPT-4o 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 77.5 to 10.2.
- The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 6.4% for GPT-4o and 98.3% for GPT-5.6 Luna.
- GPT-5.6 Luna is cheaper at $0.20 / $1.20 per million input/output tokens, against $2.50 / $10 for GPT-4o.
- GPT-5.6 Luna accepts more context: 1.05M tokens versus 128K.
Side by side
| GPT-4o | GPT-5.6 Luna | |
|---|---|---|
| Provider | OpenAI | OpenAI |
| Noometry Index | 28.5 | 54.7 |
| Released | 2024-05-13 | 2026-07-09 |
| Weights | Proprietary | Proprietary |
| Context window | 128K | 1.05M |
| Max output | 16K | 128K |
| Input $ / M tokens | $2.50 | $0.20 |
| Output $ / M tokens | $10 | $1.20 |
| Results tracked | 72 | 51 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding GPT-5.6 Luna leads
GPT-4o: 25.1 (#323), GPT-5.6 Luna: 53.9 (#25)
| Benchmark | GPT-4o | GPT-5.6 Luna |
|---|---|---|
| WeirdML | 25.1% | 60.9% |
| LMArena Coding | 1297 | 1466 |
| SWE-bench Verified | 31% | — |
| DeepSWE | — | 67.2% |
| FrontierCode | — | 39.8% |
| SWE-bench Verified (bash only) | 21.6% | — |
| Aider Polyglot | 45.3% | — |
| CursorBench | — | 35.9% |
| LMArena WebDev | — | 1517 |
| SciCode | — | 53.6% |
| GSO | 0% | — |
| BigCodeBench Instruct | 51.1% | — |
| LiveBench Coding | 51.4% | — |
| BigCodeBench Complete | 61.1% | — |
| CadEval | 26% | — |
| ALE-Bench | — | 1,667 |
| HumanEval+ | 87.2% | — |
| MBPP+ | 72.2% | — |
Agentic & Tool Use GPT-5.6 Luna leads
GPT-4o: 20.8 (#133), GPT-5.6 Luna: 35.3 (#37)
| Benchmark | GPT-4o | GPT-5.6 Luna |
|---|---|---|
| BALROG | 32.3% | 45.6% |
| GDPval | 9.9% | — |
| TheAgentCompany | 8.6% | — |
| Cybench | 12.5% | — |
| GDP.pdf | — | 22.7% |
| LMArena Search | 1006 | — |
| METR Time Horizons | 40.8% | — |
| Vending-Bench 2 | — | 4,095 |
Reasoning GPT-5.6 Luna leads
GPT-4o: 9.4 (#338), GPT-5.6 Luna: 47.5 (#38)
| Benchmark | GPT-4o | GPT-5.6 Luna |
|---|---|---|
| ARC-AGI-2 | 0% | 59.5% |
| SimpleBench | 17.8% | 46.8% |
| ARC-AGI-1 | 4.5% | 88% |
| CritPt | 0% | 20.6% |
| Chess Puzzles | 13% | 40% |
| LMArena Hard Prompts | 1281 | 1452 |
| DTBench | 64.5% | 89.1% |
| LMCA | 16.6% | 48.5% |
| Epoch Capabilities Index | 128.97 | 156.32 |
| Kagi LLM Benchmark | — | 49.1% |
| NYT Connections (extended) | — | 69.4% |
| EnigmaEval | 0.8% | — |
| LiveBench Reasoning | 55.8% | — |
| Mystery Game Puzzles | — | 21% |
| LiveBench Data Analysis | 60.9% | — |
| Surface Evolver Bench | — | 61.9% |
| ForecastBench | 57.7 | — |
| LiveBench | 55.3% | — |
Math GPT-5.6 Luna leads
GPT-4o: 10.2 (#303), GPT-5.6 Luna: 77.5 (#12)
| Benchmark | GPT-4o | GPT-5.6 Luna |
|---|---|---|
| FrontierMath (Tiers 1-3) | 0.4% | 82.1% |
| OTIS Mock AIME 2024-2025 | 6.4% | 98.3% |
| LMArena Math | 1285 | 1455 |
| FrontierMath Tier 4 | — | 61% |
| ProofBench | — | 60% |
| Omni-MATH | 29.3% | — |
| LiveBench Math | 49.5% | — |
| MATH Level 5 | 53.3% | — |
| FrontierMath (Feb 2025 set) | 0.3% | — |
Knowledge GPT-5.6 Luna leads
GPT-4o: 28.8 (#235), GPT-5.6 Luna: 58.6 (#31)
| Benchmark | GPT-4o | GPT-5.6 Luna |
|---|---|---|
| GPQA Diamond | 49.2% | 91.6% |
| SimpleQA Verified | 26% | 41% |
| LMArena Expert | 1250 | 1479 |
| Humanity's Last Exam | 2.7% | — |
| MMLU-Pro | 71.3% | — |
| Confabulations | 15.3% | — |
| Vectara Hallucination Rate | 9.6% | — |
| GPQA (HELM) | 52% | — |
| MMLU | 88.1% | — |
Multimodal GPT-5.6 Luna leads
GPT-4o: 33.9 (#85), GPT-5.6 Luna: 43.0 (#22)
| Benchmark | GPT-4o | GPT-5.6 Luna |
|---|---|---|
| LMArena Vision | 1137 | 1259 |
| Video-MME | 71.9% | — |
| GeoBench | 71% | — |
| VPCT | 40% | — |
| Blueprint-Bench 2 | — | 22.6% |
| Furniture Assembly | — | 42.5% |
| LMArena Document | — | 1457 |
| ScienceQA | 88.5% | — |
Multilingual GPT-5.6 Luna leads
GPT-4o: 43.4 (#181), GPT-5.6 Luna: 53.0 (#73)
| Benchmark | GPT-4o | GPT-5.6 Luna |
|---|---|---|
| LMArena Non-English | 1283 | 1418 |
| LMArena Chinese | 1277 | 1469 |
| LMArena French | 1303 | 1457 |
| LMArena German | 1282 | 1459 |
| LMArena Japanese | 1257 | 1416 |
| LMArena Korean | 1234 | 1414 |
| LMArena Russian | 1286 | 1428 |
| LMArena Spanish | 1292 | 1445 |
Instruction Following GPT-5.6 Luna leads
GPT-4o: 66.8 (#202), GPT-5.6 Luna: 76.0 (#48)
| Benchmark | GPT-4o | GPT-5.6 Luna |
|---|---|---|
| LMArena Instruction Following | 1278 | 1440 |
| LiveBench Instruction Following | 68.6% | — |
| IFEval | 81.7% | — |
Long Context GPT-5.6 Luna leads
GPT-4o: 39.4 (#174), GPT-5.6 Luna: 43.9 (#78)
| Benchmark | GPT-4o | GPT-5.6 Luna |
|---|---|---|
| LMArena Longer Query | 1289 | 1437 |
| Fiction.LiveBench | 66.7% | — |
Writing & Preference GPT-5.6 Luna leads
GPT-4o: 52.8 (#160), GPT-5.6 Luna: 68.3 (#27)
| Benchmark | GPT-4o | GPT-5.6 Luna |
|---|---|---|
| LMArena Text | 1300 | 1432 |
| LMArena Creative Writing | 1291 | 1397 |
| LMArena Multi-Turn | 1302 | 1433 |
| Short-Story Creative Writing | 81.8% | — |
| EQ-Bench Creative Writing | — | 1829 |
| WildBench | 82.8% | — |
| EQ-Bench 4 | — | 1156 |
| LiveBench Language | 47.6% | — |
Frequently asked questions
Is GPT-4o better than GPT-5.6 Luna?
GPT-5.6 Luna is the stronger model overall, scoring 54.7 to 28.5 on the Noometry Index.
Which is cheaper, GPT-4o or GPT-5.6 Luna?
GPT-5.6 Luna is cheaper. It lists at $0.20 per million input tokens and $1.20 per million output tokens; GPT-4o lists at $2.50 and $10.
Is GPT-4o or GPT-5.6 Luna better for coding?
GPT-5.6 Luna scores higher on coding benchmarks: 53.9 versus 25.1 in the Noometry coding category.
Which has the bigger context window?
GPT-5.6 Luna does, with 1.05M tokens against 128K.
How many benchmarks do GPT-4o and GPT-5.6 Luna share?
32 benchmarks have published results for both models. GPT-4o has 72 scored results on Noometry and GPT-5.6 Luna has 51.