Model comparison
GPT-6 Luna vs o3-pro
GPT-6 Luna is the stronger model overall, scoring 53.3 to 42.9 on the Noometry Index.
Last verified . 5 shared benchmarks.
Summary
- They share 5 benchmarks with published results for both. GPT-6 Luna scores higher in 3 categories and o3-pro in 2 categories; 4 gaps are clear of the uncertainty.
- The widest gap is in long context, where o3-pro leads 72.2 to 43.0.
- The biggest single-benchmark swing is ARC-AGI-2: 59.3% for GPT-6 Luna and 4.9% for o3-pro.
- GPT-6 Luna is cheaper at $0.10 / $0.50 per million input/output tokens, against $20 / $80 for o3-pro.
- GPT-6 Luna accepts more context: 1.05M tokens versus 200K.
Side by side
| GPT-6 Luna | o3-pro | |
|---|---|---|
| Provider | OpenAI | OpenAI |
| Noometry Index | 53.3 | 42.9 |
| Released | 2026-09-22 | 2025-06-10 |
| Weights | Proprietary | Proprietary |
| Context window | 1.05M | 200K |
| Max output | 128K | 100K |
| Input $ / M tokens | $0.10 | $20 |
| Output $ / M tokens | $0.50 | $80 |
| Results tracked | 42 | 12 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Too close to call
GPT-6 Luna: 55.5 (#25), o3-pro: 55.5 (#24)
| Benchmark | GPT-6 Luna | o3-pro |
|---|---|---|
| DeepSWE | 66.6% | — |
| FrontierCode | 42.4% | — |
| Aider Polyglot | — | 84.9% |
| LMArena WebDev | 1581 | — |
| SciCode | 54.6% | — |
| WeirdML | — | 58.2% |
| LMArena Coding | 1439 | — |
| ALE-Bench | 1,577 | — |
Agentic & Tool Use Not comparable
GPT-6 Luna: 33.3 (#54), o3-pro: —
| Benchmark | GPT-6 Luna | o3-pro |
|---|---|---|
| APEX-Agents | 44.3% | — |
| GDP.pdf | 23% | — |
Reasoning GPT-6 Luna leads
GPT-6 Luna: 48.2 (#41), o3-pro: 23.8 (#171)
| Benchmark | GPT-6 Luna | o3-pro |
|---|---|---|
| ARC-AGI-2 | 59.3% | 4.9% |
| ARC-AGI-1 | 86.7% | 59.3% |
| DTBench | 90.1% | 86.9% |
| LMCA | 44.5% | 38.5% |
| Epoch Capabilities Index | 156.28 | 147.42 |
| Kagi LLM Benchmark | — | 72.1% |
| NYT Connections (extended) | 68.7% | — |
| CritPt | 19.4% | — |
| Chess Puzzles | 31% | — |
| LMArena Hard Prompts | 1411 | — |
| Mystery Game Puzzles | 7% | — |
Math Not comparable
GPT-6 Luna: 76.1 (#15), o3-pro: —
| Benchmark | GPT-6 Luna | o3-pro |
|---|---|---|
| FrontierMath (Tiers 1-3) | 78.9% | — |
| FrontierMath Tier 4 | 56.1% | — |
| OTIS Mock AIME 2024-2025 | 98.9% | — |
| ProofBench | 64% | — |
| LMArena Math | 1416 | — |
Knowledge GPT-6 Luna leads
GPT-6 Luna: 57.0 (#41), o3-pro: 29.5 (#238)
| Benchmark | GPT-6 Luna | o3-pro |
|---|---|---|
| GPQA Diamond | 90.5% | — |
| SimpleQA Verified | 41.4% | — |
| Confabulations | — | 14.2% |
| Vectara Hallucination Rate | — | 23.3% |
| LMArena Expert | 1444 | — |
Multimodal Not comparable
GPT-6 Luna: 42.4 (#30), o3-pro: —
| Benchmark | GPT-6 Luna | o3-pro |
|---|---|---|
| LMArena Vision | 1217 | — |
| Blueprint-Bench 2 | 31.2% | — |
| Furniture Assembly | 44.2% | — |
Multilingual Not comparable
GPT-6 Luna: 50.5 (#117), o3-pro: —
| Benchmark | GPT-6 Luna | o3-pro |
|---|---|---|
| LMArena Non-English | 1386 | — |
| LMArena Chinese | 1433 | — |
| LMArena French | 1420 | — |
| LMArena German | 1369 | — |
| LMArena Japanese | 1369 | — |
| LMArena Korean | 1360 | — |
| LMArena Russian | 1394 | — |
| LMArena Spanish | 1393 | — |
Instruction Following Not comparable
GPT-6 Luna: 74.3 (#99), o3-pro: —
| Benchmark | GPT-6 Luna | o3-pro |
|---|---|---|
| LMArena Instruction Following | 1409 | — |
Long Context o3-pro leads
GPT-6 Luna: 43.0 (#111), o3-pro: 72.2 (#1)
| Benchmark | GPT-6 Luna | o3-pro |
|---|---|---|
| Fiction.LiveBench | — | 97.2% |
| LMArena Longer Query | 1409 | — |
Writing & Preference GPT-6 Luna leads
GPT-6 Luna: 58.3 (#119), o3-pro: 57.1 (#133)
| Benchmark | GPT-6 Luna | o3-pro |
|---|---|---|
| LMArena Text | 1391 | — |
| LMArena Creative Writing | 1363 | — |
| Short-Story Creative Writing | — | 84.4% |
| LMArena Multi-Turn | 1396 | — |
Frequently asked questions
Is GPT-6 Luna better than o3-pro?
GPT-6 Luna is the stronger model overall, scoring 53.3 to 42.9 on the Noometry Index.
Which is cheaper, GPT-6 Luna or o3-pro?
GPT-6 Luna is cheaper. It lists at $0.10 per million input tokens and $0.50 per million output tokens; o3-pro lists at $20 and $80.
Is GPT-6 Luna or o3-pro better for coding?
They score almost the same on coding (55.5 vs 55.5); test both on your own repository before choosing.
Which has the bigger context window?
GPT-6 Luna does, with 1.05M tokens against 200K.
How many benchmarks do GPT-6 Luna and o3-pro share?
5 benchmarks have published results for both models. GPT-6 Luna has 42 scored results on Noometry and o3-pro has 12.