# GPT-6 Luna vs o1-mini

> GPT-6 Luna is the stronger model overall, scoring 53.3 to 34.0 on the Noometry Index.

- Canonical page: https://noometry.com/compare/gpt-6-luna-vs-o1-mini
- Last updated: 2026-10-10
- Shared benchmarks: 22

## Summary

- They share 22 benchmarks with published results for both. GPT-6 Luna scores higher in 9 categories and o1-mini in 0 categories; 9 gaps are clear of the uncertainty.
- The widest gap is in math, where GPT-6 Luna leads 76.1 to 35.4.
- The biggest single-benchmark swing is ARC-AGI-1: 86.7% for GPT-6 Luna and 14% for o1-mini.

## Snapshot

| | GPT-6 Luna | o1-mini |
|---|---|---|
| Provider | OpenAI | OpenAI |
| Noometry Index | 53.3 | 34.0 |
| Rank | 36 | 235 |
| Context | 1.05M | — |
| Input $/M | $0.10 | — |
| Output $/M | $0.50 | — |
| Weights | Proprietary | Proprietary |

## Coding

- GPT-6 Luna: 55.5 (#25)
- o1-mini: 35.5 (#224)

| Benchmark | GPT-6 Luna | o1-mini |
|---|---|---|
| LMArena Coding | 1439 | 1362 |
| DeepSWE | 66.6% | — |
| FrontierCode | 42.4% | — |
| Aider Polyglot | — | 32.9% |
| LMArena WebDev | 1581 | — |
| SciCode | 54.6% | — |
| WeirdML | — | 36.3% |
| LiveBench Coding | — | 48% |
| ALE-Bench | 1,577 | — |
| HumanEval+ | — | 89% |
| MBPP+ | — | 78.8% |

## Agentic & Tool Use

- GPT-6 Luna: 33.3 (#54)
- o1-mini: 24.6 (#118)

| Benchmark | GPT-6 Luna | o1-mini |
|---|---|---|
| APEX-Agents | 44.3% | — |
| Cybench | — | 10% |
| GDP.pdf | 23% | — |

## Reasoning

- GPT-6 Luna: 48.2 (#41)
- o1-mini: 8.8 (#346)

| Benchmark | GPT-6 Luna | o1-mini |
|---|---|---|
| ARC-AGI-2 | 59.3% | 0.8% |
| ARC-AGI-1 | 86.7% | 14% |
| LMArena Hard Prompts | 1411 | 1333 |
| Epoch Capabilities Index | 156.28 | 135.82 |
| SimpleBench | — | 18.1% |
| NYT Connections (extended) | 68.7% | — |
| CritPt | 19.4% | — |
| Chess Puzzles | 31% | — |
| LiveBench Reasoning | — | 72.3% |
| Mystery Game Puzzles | 7% | — |
| DTBench | 90.1% | — |
| LiveBench Data Analysis | — | 57.9% |
| LMCA | 44.5% | — |
| LiveBench | — | 57.8% |

## Math

- GPT-6 Luna: 76.1 (#15)
- o1-mini: 35.4 (#186)

| Benchmark | GPT-6 Luna | o1-mini |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 98.9% | 46.9% |
| LMArena Math | 1416 | 1358 |
| FrontierMath (Tiers 1-3) | 78.9% | — |
| FrontierMath Tier 4 | 56.1% | — |
| ProofBench | 64% | — |
| LiveBench Math | — | 62% |
| MATH Level 5 | — | 89.2% |
| FrontierMath (Feb 2025 set) | — | 1.7% |

## Knowledge

- GPT-6 Luna: 57.0 (#41)
- o1-mini: 34.9 (#192)

| Benchmark | GPT-6 Luna | o1-mini |
|---|---|---|
| GPQA Diamond | 90.5% | 62.4% |
| LMArena Expert | 1444 | 1316 |
| SimpleQA Verified | 41.4% | — |
| Confabulations | — | 18.6% |

## Multimodal

- GPT-6 Luna: 42.4 (#30)
- o1-mini: —

| Benchmark | GPT-6 Luna | o1-mini |
|---|---|---|
| LMArena Vision | 1217 | — |
| Blueprint-Bench 2 | 31.2% | — |
| Furniture Assembly | 44.2% | — |

## Multilingual

- GPT-6 Luna: 50.5 (#117)
- o1-mini: 43.6 (#182)

| Benchmark | GPT-6 Luna | o1-mini |
|---|---|---|
| LMArena Non-English | 1386 | 1289 |
| LMArena Chinese | 1433 | 1314 |
| LMArena French | 1420 | 1293 |
| LMArena German | 1369 | 1278 |
| LMArena Japanese | 1369 | 1245 |
| LMArena Korean | 1360 | 1223 |
| LMArena Russian | 1394 | 1283 |
| LMArena Spanish | 1393 | 1303 |

## Instruction Following

- GPT-6 Luna: 74.3 (#99)
- o1-mini: 66.7 (#206)

| Benchmark | GPT-6 Luna | o1-mini |
|---|---|---|
| LMArena Instruction Following | 1409 | 1304 |
| LiveBench Instruction Following | — | 65.4% |

## Long Context

- GPT-6 Luna: 43.0 (#111)
- o1-mini: 40.1 (#161)

| Benchmark | GPT-6 Luna | o1-mini |
|---|---|---|
| LMArena Longer Query | 1409 | 1320 |

## Writing & Preference

- GPT-6 Luna: 58.3 (#119)
- o1-mini: 48.4 (#202)

| Benchmark | GPT-6 Luna | o1-mini |
|---|---|---|
| LMArena Text | 1391 | 1317 |
| LMArena Creative Writing | 1363 | 1244 |
| LMArena Multi-Turn | 1396 | 1314 |
| Short-Story Creative Writing | — | 64.9% |
| LiveBench Language | — | 40.9% |

## FAQ

### Is GPT-6 Luna better than o1-mini?

GPT-6 Luna is the stronger model overall, scoring 53.3 to 34.0 on the Noometry Index.

### Is GPT-6 Luna or o1-mini better for coding?

GPT-6 Luna scores higher on coding benchmarks: 55.5 versus 35.5 in the Noometry coding category.

### How many benchmarks do GPT-6 Luna and o1-mini share?

22 benchmarks have published results for both models. GPT-6 Luna has 42 scored results on Noometry and o1-mini has 39.