# Claude Opus 4 vs GPT-6 Luna

> GPT-6 Luna is the stronger model overall, scoring 53.3 to 43.1 on the Noometry Index.

- Canonical page: https://noometry.com/compare/claude-opus-4-vs-gpt-6-luna
- Last updated: 2026-10-12
- Shared benchmarks: 26

## Summary

- They share 26 benchmarks with published results for both. Claude Opus 4 scores higher in 3 categories and GPT-6 Luna in 7 categories; 10 gaps are clear of the uncertainty.
- The widest gap is in math, where GPT-6 Luna leads 76.1 to 42.0.
- The biggest single-benchmark swing is ARC-AGI-1: 35.7% for Claude Opus 4 and 86.7% for GPT-6 Luna.
- GPT-6 Luna is cheaper at $0.10 / $0.50 per million input/output tokens, against $15 / $75 for Claude Opus 4.
- GPT-6 Luna accepts more context: 1.05M tokens versus 200K.

## Snapshot

| | Claude Opus 4 | GPT-6 Luna |
|---|---|---|
| Provider | Anthropic | OpenAI |
| Noometry Index | 43.1 | 53.3 |
| Rank | 100 | 36 |
| Context | 200K | 1.05M |
| Input $/M | $15 | $0.10 |
| Output $/M | $75 | $0.50 |
| Weights | Proprietary | Proprietary |

## Coding

- Claude Opus 4: 47.2 (#62)
- GPT-6 Luna: 55.5 (#25)

| Benchmark | Claude Opus 4 | GPT-6 Luna |
|---|---|---|
| LMArena Coding | 1442 | 1439 |
| SWE-bench Verified | 70.7% | — |
| DeepSWE | — | 66.6% |
| FrontierCode | — | 42.4% |
| SWE-bench Verified (bash only) | 67.6% | — |
| Aider Polyglot | 72% | — |
| LMArena WebDev | — | 1581 |
| SciCode | — | 54.6% |
| GSO | 6.9% | — |
| WeirdML | 43.7% | — |
| ALE-Bench | — | 1,577 |
| AlgoTune | 1.33 | — |

## Agentic & Tool Use

- Claude Opus 4: 34.8 (#42)
- GPT-6 Luna: 33.3 (#54)

| Benchmark | Claude Opus 4 | GPT-6 Luna |
|---|---|---|
| APEX-Agents | — | 44.3% |
| Cybench | 38% | — |
| DeepResearch Bench | 46.8% | — |
| GDP.pdf | — | 23% |
| LMArena Search | 1127 | — |
| METR Time Horizons | 63.9% | — |

## Reasoning

- Claude Opus 4: 27.3 (#121)
- GPT-6 Luna: 48.2 (#41)

| Benchmark | Claude Opus 4 | GPT-6 Luna |
|---|---|---|
| ARC-AGI-2 | 8.6% | 59.3% |
| ARC-AGI-1 | 35.7% | 86.7% |
| CritPt | 0.3% | 19.4% |
| LMArena Hard Prompts | 1399 | 1411 |
| DTBench | 81.6% | 90.1% |
| LMCA | 37.4% | 44.5% |
| Epoch Capabilities Index | 142.67 | 156.28 |
| SimpleBench | 58.8% | — |
| Kagi LLM Benchmark | 74.3% | — |
| NYT Connections (extended) | — | 68.7% |
| Chess Puzzles | — | 31% |
| EnigmaEval | 5.6% | — |
| Mystery Game Puzzles | — | 7% |
| ForecastBench | 61.1 | — |

## Math

- Claude Opus 4: 42.0 (#86)
- GPT-6 Luna: 76.1 (#15)

| Benchmark | Claude Opus 4 | GPT-6 Luna |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 64.4% | 98.9% |
| LMArena Math | 1390 | 1416 |
| FrontierMath (Tiers 1-3) | — | 78.9% |
| FrontierMath Tier 4 | — | 56.1% |
| ProofBench | — | 64% |
| Omni-MATH | 61.6% | — |
| MATH Level 5 | 85% | — |
| FrontierMath (Feb 2025 set) | 4.5% | — |
| FrontierMath Tier 4 (v1) | 4.2% | — |

## Knowledge

- Claude Opus 4: 44.0 (#88)
- GPT-6 Luna: 57.0 (#41)

| Benchmark | Claude Opus 4 | GPT-6 Luna |
|---|---|---|
| GPQA Diamond | 76.3% | 90.5% |
| LMArena Expert | 1386 | 1444 |
| Humanity's Last Exam | 10.7% | — |
| SimpleQA Verified | — | 41.4% |
| MMLU-Pro | 87.5% | — |
| Confabulations | 15.9% | — |
| Vectara Hallucination Rate | 12% | — |
| GPQA (HELM) | 70.8% | — |

## Multimodal

- Claude Opus 4: 31.5 (#106)
- GPT-6 Luna: 42.4 (#30)

| Benchmark | Claude Opus 4 | GPT-6 Luna |
|---|---|---|
| LMArena Vision | 1192 | 1217 |
| GeoBench | 49% | — |
| VPCT | 38% | — |
| Blueprint-Bench 2 | — | 31.2% |
| Furniture Assembly | — | 44.2% |

## Multilingual

- Claude Opus 4: 48.8 (#138)
- GPT-6 Luna: 50.5 (#117)

| Benchmark | Claude Opus 4 | GPT-6 Luna |
|---|---|---|
| LMArena Non-English | 1362 | 1386 |
| LMArena Chinese | 1386 | 1433 |
| LMArena French | 1372 | 1420 |
| LMArena German | 1391 | 1369 |
| LMArena Japanese | 1331 | 1369 |
| LMArena Korean | 1321 | 1360 |
| LMArena Russian | 1392 | 1394 |
| LMArena Spanish | 1389 | 1393 |

## Instruction Following

- Claude Opus 4: 77.1 (#28)
- GPT-6 Luna: 74.3 (#99)

| Benchmark | Claude Opus 4 | GPT-6 Luna |
|---|---|---|
| LMArena Instruction Following | 1406 | 1409 |
| IFEval | 91.8% | — |

## Long Context

- Claude Opus 4: 39.6 (#172)
- GPT-6 Luna: 43.0 (#111)

| Benchmark | Claude Opus 4 | GPT-6 Luna |
|---|---|---|
| LMArena Longer Query | 1422 | 1409 |
| Fiction.LiveBench | 61.1% | — |

## Writing & Preference

- Claude Opus 4: 61.2 (#89)
- GPT-6 Luna: 58.3 (#119)

| Benchmark | Claude Opus 4 | GPT-6 Luna |
|---|---|---|
| LMArena Text | 1377 | 1391 |
| LMArena Creative Writing | 1387 | 1363 |
| LMArena Multi-Turn | 1396 | 1396 |
| Short-Story Creative Writing | 83.6% | — |
| EQ-Bench Creative Writing | 1580 | — |
| WildBench | 85.2% | — |

## FAQ

### Is Claude Opus 4 better than GPT-6 Luna?

GPT-6 Luna is the stronger model overall, scoring 53.3 to 43.1 on the Noometry Index.

### Which is cheaper, Claude Opus 4 or GPT-6 Luna?

GPT-6 Luna is cheaper. It lists at $0.10 per million input tokens and $0.50 per million output tokens; Claude Opus 4 lists at $15 and $75.

### Is Claude Opus 4 or GPT-6 Luna better for coding?

GPT-6 Luna scores higher on coding benchmarks: 55.5 versus 47.2 in the Noometry coding category.

### Which has the bigger context window?

GPT-6 Luna does, with 1.05M tokens against 200K.

### How many benchmarks do Claude Opus 4 and GPT-6 Luna share?

26 benchmarks have published results for both models. Claude Opus 4 has 56 scored results on Noometry and GPT-6 Luna has 42.