Model comparison
GPT-5.6 Luna vs o3-pro
GPT-5.6 Luna is the stronger model overall, scoring 54.7 to 42.7 on the Noometry Index.
Last verified . 7 shared benchmarks.
Summary
- They share 7 benchmarks with published results for both. GPT-5.6 Luna scores higher in 3 categories and o3-pro in 2 categories; 5 gaps are clear of the uncertainty.
- The widest gap is in knowledge, where GPT-5.6 Luna leads 58.6 to 29.0.
- The biggest single-benchmark swing is ARC-AGI-2: 59.5% for GPT-5.6 Luna and 4.9% for o3-pro.
- GPT-5.6 Luna is cheaper at $0.20 / $1.20 per million input/output tokens, against $20 / $80 for o3-pro.
- GPT-5.6 Luna accepts more context: 1.05M tokens versus 200K.
Side by side
| GPT-5.6 Luna | o3-pro | |
|---|---|---|
| Provider | OpenAI | OpenAI |
| Noometry Index | 54.7 | 42.7 |
| Released | 2026-07-09 | 2025-06-10 |
| Weights | Proprietary | Proprietary |
| Context window | 1.05M | 200K |
| Max output | 128K | 100K |
| Input $ / M tokens | $0.20 | $20 |
| Output $ / M tokens | $1.20 | $80 |
| Results tracked | 51 | 12 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding o3-pro leads
GPT-5.6 Luna: 53.9 (#25), o3-pro: 55.1 (#22)
| Benchmark | GPT-5.6 Luna | o3-pro |
|---|---|---|
| WeirdML | 60.9% | 58.2% |
| DeepSWE | 67.2% | — |
| FrontierCode | 39.8% | — |
| Aider Polyglot | — | 84.9% |
| CursorBench | 35.9% | — |
| LMArena WebDev | 1517 | — |
| SciCode | 53.6% | — |
| LMArena Coding | 1466 | — |
| ALE-Bench | 1,667 | — |
Agentic & Tool Use Not comparable
GPT-5.6 Luna: 35.3 (#37), o3-pro: —
| Benchmark | GPT-5.6 Luna | o3-pro |
|---|---|---|
| BALROG | 45.6% | — |
| GDP.pdf | 22.7% | — |
| Vending-Bench 2 | 4,095 | — |
Reasoning GPT-5.6 Luna leads
GPT-5.6 Luna: 47.5 (#38), o3-pro: 24.1 (#152)
| Benchmark | GPT-5.6 Luna | o3-pro |
|---|---|---|
| ARC-AGI-2 | 59.5% | 4.9% |
| Kagi LLM Benchmark | 49.1% | 72.1% |
| ARC-AGI-1 | 88% | 59.3% |
| DTBench | 89.1% | 86.9% |
| LMCA | 48.5% | 38.5% |
| Epoch Capabilities Index | 156.32 | 147.43 |
| SimpleBench | 46.8% | — |
| NYT Connections (extended) | 69.4% | — |
| CritPt | 20.6% | — |
| Chess Puzzles | 40% | — |
| LMArena Hard Prompts | 1452 | — |
| Mystery Game Puzzles | 21% | — |
| Surface Evolver Bench | 61.9% | — |
Math Not comparable
GPT-5.6 Luna: 77.5 (#12), o3-pro: —
| Benchmark | GPT-5.6 Luna | o3-pro |
|---|---|---|
| FrontierMath (Tiers 1-3) | 82.1% | — |
| FrontierMath Tier 4 | 61% | — |
| OTIS Mock AIME 2024-2025 | 98.3% | — |
| ProofBench | 60% | — |
| LMArena Math | 1455 | — |
Knowledge GPT-5.6 Luna leads
GPT-5.6 Luna: 58.6 (#31), o3-pro: 29.0 (#234)
| Benchmark | GPT-5.6 Luna | o3-pro |
|---|---|---|
| GPQA Diamond | 91.6% | — |
| SimpleQA Verified | 41% | — |
| Confabulations | — | 14.2% |
| Vectara Hallucination Rate | — | 23.3% |
| LMArena Expert | 1479 | — |
Multimodal Not comparable
GPT-5.6 Luna: 43.0 (#22), o3-pro: —
| Benchmark | GPT-5.6 Luna | o3-pro |
|---|---|---|
| LMArena Vision | 1259 | — |
| Blueprint-Bench 2 | 22.6% | — |
| Furniture Assembly | 42.5% | — |
| LMArena Document | 1457 | — |
Multilingual Not comparable
GPT-5.6 Luna: 53.0 (#73), o3-pro: —
| Benchmark | GPT-5.6 Luna | o3-pro |
|---|---|---|
| LMArena Non-English | 1418 | — |
| LMArena Chinese | 1469 | — |
| LMArena French | 1457 | — |
| LMArena German | 1459 | — |
| LMArena Japanese | 1416 | — |
| LMArena Korean | 1414 | — |
| LMArena Russian | 1428 | — |
| LMArena Spanish | 1445 | — |
Instruction Following Not comparable
GPT-5.6 Luna: 76.0 (#48), o3-pro: —
| Benchmark | GPT-5.6 Luna | o3-pro |
|---|---|---|
| LMArena Instruction Following | 1440 | — |
Long Context o3-pro leads
GPT-5.6 Luna: 43.9 (#78), o3-pro: 72.1 (#1)
| Benchmark | GPT-5.6 Luna | o3-pro |
|---|---|---|
| Fiction.LiveBench | — | 97.2% |
| LMArena Longer Query | 1437 | — |
Writing & Preference GPT-5.6 Luna leads
GPT-5.6 Luna: 68.3 (#27), o3-pro: 57.1 (#128)
| Benchmark | GPT-5.6 Luna | o3-pro |
|---|---|---|
| LMArena Text | 1432 | — |
| LMArena Creative Writing | 1397 | — |
| Short-Story Creative Writing | — | 84.4% |
| EQ-Bench Creative Writing | 1829 | — |
| EQ-Bench 4 | 1156 | — |
| LMArena Multi-Turn | 1433 | — |
Frequently asked questions
Is GPT-5.6 Luna better than o3-pro?
GPT-5.6 Luna is the stronger model overall, scoring 54.7 to 42.7 on the Noometry Index.
Which is cheaper, GPT-5.6 Luna or o3-pro?
GPT-5.6 Luna is cheaper. It lists at $0.20 per million input tokens and $1.20 per million output tokens; o3-pro lists at $20 and $80.
Is GPT-5.6 Luna or o3-pro better for coding?
o3-pro scores higher on coding benchmarks: 55.1 versus 53.9 in the Noometry coding category.
Which has the bigger context window?
GPT-5.6 Luna does, with 1.05M tokens against 200K.
How many benchmarks do GPT-5.6 Luna and o3-pro share?
7 benchmarks have published results for both models. GPT-5.6 Luna has 51 scored results on Noometry and o3-pro has 12.