# Claude Opus 5 vs GPT-4.5

> Claude Opus 5 is the stronger model overall, scoring 67.7 to 36.9 on the Noometry Index.

- Canonical page: https://noometry.com/compare/claude-opus-5-vs-gpt-4-5
- Last updated: 2026-10-10
- Shared benchmarks: 25

## Summary

- They share 25 benchmarks with published results for both. Claude Opus 5 scores higher in 10 categories and GPT-4.5 in 0 categories; 10 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where Claude Opus 5 leads 77.1 to 14.1.
- The biggest single-benchmark swing is ARC-AGI-2: 90.4% for Claude Opus 5 and 0.8% for GPT-4.5.

## Snapshot

| | Claude Opus 5 | GPT-4.5 |
|---|---|---|
| Provider | Anthropic | OpenAI |
| Noometry Index | 67.7 | 36.9 |
| Rank | 4 | 200 |
| Context | 1M | — |
| Input $/M | $5 | — |
| Output $/M | $25 | — |
| Weights | Proprietary | Proprietary |

## Coding

- Claude Opus 5: 67.0 (#5)
- GPT-4.5: 41.8 (#102)

| Benchmark | Claude Opus 5 | GPT-4.5 |
|---|---|---|
| WeirdML | 91.8% | 39.4% |
| LMArena Coding | 1533 | 1397 |
| DeepSWE | 73.6% | — |
| FrontierCode | 53.4% | — |
| Aider Polyglot | — | 44.9% |
| CursorBench | 46.6% | — |
| LMArena WebDev | 1694 | — |
| FrontierSWE | 52% | — |
| SciCode | 56.4% | — |
| LiveBench Coding | — | 75.2% |
| ALE-Bench | 2,165 | — |

## Agentic & Tool Use

- Claude Opus 5: 55.5 (#1)
- GPT-4.5: 27.6 (#92)

| Benchmark | Claude Opus 5 | GPT-4.5 |
|---|---|---|
| APEX-Agents | 65.8% | — |
| OSWorld 2.0 | 31.4% | — |
| τ²-bench Banking | 48.7% | — |
| Cybench | — | 17.5% |
| PostTrainBench | 35% | — |
| BALROG | 63.4% | — |
| GBAEval | 79.6% | — |
| GDP.pdf | 24% | — |
| Vending-Bench 2 | 11,182 | — |

## Reasoning

- Claude Opus 5: 77.1 (#3)
- GPT-4.5: 14.1 (#320)

| Benchmark | Claude Opus 5 | GPT-4.5 |
|---|---|---|
| ARC-AGI-2 | 90.4% | 0.8% |
| SimpleBench | 80.6% | 34.5% |
| ARC-AGI-1 | 97.5% | 10.3% |
| LMArena Hard Prompts | 1525 | 1404 |
| Epoch Capabilities Index | 162.67 | 136.74 |
| NYT Connections (extended) | 94.3% | — |
| CritPt | 29.1% | — |
| Chess Puzzles | 42% | — |
| EnigmaEval | — | 3.2% |
| EBR-Bench | 45.7% | — |
| LiveBench Reasoning | — | 71.1% |
| Mystery Game Puzzles | 59% | — |
| DTBench | 97.9% | — |
| LiveBench Data Analysis | — | 64.3% |
| LMCA | 64.5% | — |
| Bench to the Future 3 | 0.12 | — |
| ForecastBench | — | 61.7 |
| LiveBench | — | 69% |

## Math

- Claude Opus 5: 86.1 (#6)
- GPT-4.5: 31.0 (#200)

| Benchmark | Claude Opus 5 | GPT-4.5 |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 98.9% | 37.8% |
| LMArena Math | 1533 | 1412 |
| FrontierMath (Tiers 1-3) | 85.6% | — |
| FrontierMath Tier 4 | 73.2% | — |
| ProofBench | 99% | — |
| LiveBench Math | — | 69.3% |
| MATH Level 5 | — | 78.6% |

## Knowledge

- Claude Opus 5: 66.8 (#8)
- GPT-4.5: 32.4 (#205)

| Benchmark | Claude Opus 5 | GPT-4.5 |
|---|---|---|
| GPQA Diamond | 93.9% | 68.7% |
| LMArena Expert | 1558 | 1394 |
| Humanity's Last Exam | — | 5.4% |
| SimpleQA Verified | 59.9% | — |
| Confabulations | — | 13.6% |

## Multimodal

- Claude Opus 5: 51.1 (#6)
- GPT-4.5: 37.0 (#68)

| Benchmark | Claude Opus 5 | GPT-4.5 |
|---|---|---|
| LMArena Vision | 1321 | 1195 |
| VPCT | — | 45% |
| Blueprint-Bench 2 | 30.4% | — |
| Furniture Assembly | 60.8% | — |
| LMArena Document | 1516 | — |

## Multilingual

- Claude Opus 5: 58.8 (#3)
- GPT-4.5: 52.7 (#78)

| Benchmark | Claude Opus 5 | GPT-4.5 |
|---|---|---|
| LMArena Non-English | 1500 | 1413 |
| LMArena Chinese | 1575 | 1421 |
| LMArena French | 1522 | 1417 |
| LMArena German | 1522 | 1457 |
| LMArena Japanese | 1510 | 1416 |
| LMArena Korean | 1526 | 1392 |
| LMArena Russian | 1507 | 1419 |
| LMArena Spanish | 1519 | — |

## Instruction Following

- Claude Opus 5: 79.3 (#5)
- GPT-4.5: 72.8 (#129)

| Benchmark | Claude Opus 5 | GPT-4.5 |
|---|---|---|
| LMArena Instruction Following | 1516 | 1404 |
| LiveBench Instruction Following | — | 72.3% |

## Long Context

- Claude Opus 5: 46.5 (#20)
- GPT-4.5: 40.4 (#150)

| Benchmark | Claude Opus 5 | GPT-4.5 |
|---|---|---|
| LMArena Longer Query | 1516 | 1406 |
| Fiction.LiveBench | — | 63.9% |

## Writing & Preference

- Claude Opus 5: 79.3 (#2)
- GPT-4.5: 57.1 (#129)

| Benchmark | Claude Opus 5 | GPT-4.5 |
|---|---|---|
| LMArena Text | 1506 | 1417 |
| LMArena Creative Writing | 1491 | 1394 |
| EQ-Bench Creative Writing | 2133 | 1258 |
| LMArena Multi-Turn | 1496 | 1444 |
| Short-Story Creative Writing | — | 75.6% |
| EQ-Bench 4 | 1385 | — |
| LiveBench Language | — | 61.5% |

## FAQ

### Is Claude Opus 5 better than GPT-4.5?

Claude Opus 5 is the stronger model overall, scoring 67.7 to 36.9 on the Noometry Index.

### Is Claude Opus 5 or GPT-4.5 better for coding?

Claude Opus 5 scores higher on coding benchmarks: 67.0 versus 41.8 in the Noometry coding category.

### How many benchmarks do Claude Opus 5 and GPT-4.5 share?

25 benchmarks have published results for both models. Claude Opus 5 has 57 scored results on Noometry and GPT-4.5 has 42.