# Claude Opus 4 vs Claude Opus 4.5

> Claude Opus 4.5 is the stronger model overall, scoring 50.5 to 42.9 on the Noometry Index.

- Canonical page: https://noometry.com/compare/claude-opus-4-vs-claude-opus-4-5
- Last updated: 2026-10-10
- Shared benchmarks: 44

## Summary

- They share 44 benchmarks with published results for both. Claude Opus 4 scores higher in 1 category and Claude Opus 4.5 in 9 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where Claude Opus 4.5 leads 42.5 to 27.4.
- The biggest single-benchmark swing is ARC-AGI-1: 35.7% for Claude Opus 4 and 80% for Claude Opus 4.5.
- Claude Opus 4.5 is cheaper at $5 / $25 per million input/output tokens, against $15 / $75 for Claude Opus 4.

## Snapshot

| | Claude Opus 4 | Claude Opus 4.5 |
|---|---|---|
| Provider | Anthropic | Anthropic |
| Noometry Index | 42.9 | 50.5 |
| Rank | 93 | 39 |
| Context | 200K | 200K |
| Input $/M | $15 | $5 |
| Output $/M | $75 | $25 |
| Weights | Proprietary | Proprietary |

## Coding

- Claude Opus 4: 47.8 (#55)
- Claude Opus 4.5: 55.4 (#21)

| Benchmark | Claude Opus 4 | Claude Opus 4.5 |
|---|---|---|
| SWE-bench Verified | 70.7% | 76.7% |
| SWE-bench Verified (bash only) | 67.6% | 76.8% |
| GSO | 6.9% | 26.5% |
| WeirdML | 43.7% | 63.7% |
| LMArena Coding | 1442 | 1504 |
| AlgoTune | 1.33 | 1.77 |
| Aider Polyglot | 72% | — |
| LMArena WebDev | — | 1493 |
| SWE-bench Multilingual | — | 70.7% |
| ALE-Bench | — | 1,025 |

## Agentic & Tool Use

- Claude Opus 4: 34.5 (#39)
- Claude Opus 4.5: 47.2 (#11)

| Benchmark | Claude Opus 4 | Claude Opus 4.5 |
|---|---|---|
| Cybench | 38% | 82% |
| DeepResearch Bench | 46.8% | 54.8% |
| LMArena Search | 1127 | 1180 |
| METR Time Horizons | 63.9% | 75% |
| Terminal-Bench | — | 63.1% |
| Berkeley Function Calling Leaderboard | — | 77.5% |
| GDPval | — | 45.5% |
| Remote Labor Index | — | 3.8% |
| τ²-bench Airline | — | 84% |
| τ²-bench Banking | — | 24.7% |
| τ²-bench Retail | — | 79.6% |
| τ²-bench Telecom | — | 92.3% |
| OSWorld | — | 66.3% |
| BALROG | — | 43.5% |
| Vending-Bench 2 | — | 4,967 |

## Reasoning

- Claude Opus 4: 27.4 (#104)
- Claude Opus 4.5: 42.5 (#43)

| Benchmark | Claude Opus 4 | Claude Opus 4.5 |
|---|---|---|
| ARC-AGI-2 | 8.6% | 37.6% |
| SimpleBench | 58.8% | 62% |
| Kagi LLM Benchmark | 74.3% | 80.2% |
| ARC-AGI-1 | 35.7% | 80% |
| EnigmaEval | 5.6% | 11.9% |
| LMArena Hard Prompts | 1399 | 1476 |
| DTBench | 81.6% | 89.9% |
| LMCA | 37.4% | 44.5% |
| Epoch Capabilities Index | 142.67 | 150.1 |
| ForecastBench | 61.1 | 60.7 |
| NYT Connections (extended) | — | 52.5% |
| CritPt | 0.3% | — |
| Chess Puzzles | — | 12% |
| EBR-Bench | — | 14.3% |
| Mystery Game Puzzles | — | 22% |

## Math

- Claude Opus 4: 40.2 (#80)
- Claude Opus 4.5: 38.5 (#85)

| Benchmark | Claude Opus 4 | Claude Opus 4.5 |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 64.4% | 86.1% |
| LMArena Math | 1391 | 1463 |
| FrontierMath (Feb 2025 set) | 4.5% | 20.7% |
| FrontierMath Tier 4 (v1) | 4.2% | 4.2% |
| FrontierMath (Tiers 1-3) | — | 34.4% |
| FrontierMath Tier 4 | — | 4.9% |
| ProofBench | — | 36% |
| Omni-MATH | 61.6% | — |
| MATH Level 5 | 85% | — |

## Knowledge

- Claude Opus 4: 43.9 (#82)
- Claude Opus 4.5: 56.4 (#38)

| Benchmark | Claude Opus 4 | Claude Opus 4.5 |
|---|---|---|
| GPQA Diamond | 76.3% | 86% |
| Humanity's Last Exam | 10.7% | 25.2% |
| Vectara Hallucination Rate | 12% | 10.9% |
| LMArena Expert | 1384 | 1488 |
| SimpleQA Verified | — | 45.7% |
| MMLU-Pro | 87.5% | — |
| Confabulations | 15.9% | — |
| GPQA (HELM) | 70.8% | — |

## Multimodal

- Claude Opus 4: 30.9 (#101)
- Claude Opus 4.5: 31.1 (#99)

| Benchmark | Claude Opus 4 | Claude Opus 4.5 |
|---|---|---|
| GeoBench | 49% | 75% |
| VPCT | 38% | 40% |
| LMArena Vision | 1192 | — |
| Furniture Assembly | — | 28.3% |
| LMArena Document | — | 1462 |

## Multilingual

- Claude Opus 4: 49.0 (#133)
- Claude Opus 4.5: 54.5 (#45)

| Benchmark | Claude Opus 4 | Claude Opus 4.5 |
|---|---|---|
| LMArena Non-English | 1362 | 1438 |
| LMArena Chinese | 1386 | 1470 |
| LMArena French | 1370 | 1471 |
| LMArena German | 1391 | 1449 |
| LMArena Japanese | 1332 | 1415 |
| LMArena Korean | 1318 | 1424 |
| LMArena Russian | 1392 | 1448 |
| LMArena Spanish | 1388 | 1455 |

## Instruction Following

- Claude Opus 4: 77.3 (#26)
- Claude Opus 4.5: 77.7 (#20)

| Benchmark | Claude Opus 4 | Claude Opus 4.5 |
|---|---|---|
| LMArena Instruction Following | 1406 | 1478 |
| IFEval | 91.8% | — |

## Long Context

- Claude Opus 4: 39.5 (#167)
- Claude Opus 4.5: 46.4 (#21)

| Benchmark | Claude Opus 4 | Claude Opus 4.5 |
|---|---|---|
| LMArena Longer Query | 1422 | 1480 |
| Fiction.LiveBench | 61.1% | — |
| CL-bench | — | 21.1% |

## Writing & Preference

- Claude Opus 4: 61.4 (#85)
- Claude Opus 4.5: 68.2 (#28)

| Benchmark | Claude Opus 4 | Claude Opus 4.5 |
|---|---|---|
| LMArena Text | 1377 | 1451 |
| LMArena Creative Writing | 1386 | 1445 |
| EQ-Bench Creative Writing | 1580 | 1687 |
| LMArena Multi-Turn | 1396 | 1466 |
| Short-Story Creative Writing | 83.6% | — |
| WildBench | 85.2% | — |

## FAQ

### Is Claude Opus 4 better than Claude Opus 4.5?

Claude Opus 4.5 is the stronger model overall, scoring 50.5 to 42.9 on the Noometry Index.

### Which is cheaper, Claude Opus 4 or Claude Opus 4.5?

Claude Opus 4.5 is cheaper. It lists at $5 per million input tokens and $25 per million output tokens; Claude Opus 4 lists at $15 and $75.

### Is Claude Opus 4 or Claude Opus 4.5 better for coding?

Claude Opus 4.5 scores higher on coding benchmarks: 55.4 versus 47.8 in the Noometry coding category.

### Which has the bigger context window?

Both accept 200K tokens.

### How many benchmarks do Claude Opus 4 and Claude Opus 4.5 share?

44 benchmarks have published results for both models. Claude Opus 4 has 56 scored results on Noometry and Claude Opus 4.5 has 69.