# Claude Opus 5 vs Claude Opus 5.5

> Claude Opus 5 and Claude Opus 5.5 score almost the same on the Noometry Index (67.8 vs 68.6), so choose on price, context window or the category you care about most.

- Canonical page: https://noometry.com/compare/claude-opus-5-vs-claude-opus-5-5
- Last updated: 2026-10-10
- Shared benchmarks: 42

## Summary

- They share 42 benchmarks with published results for both. Claude Opus 5 scores higher in 3 categories and Claude Opus 5.5 in 7 categories; 5 gaps are clear of the uncertainty.
- The widest gap is in agentic & tool use, where Claude Opus 5 leads 55.6 to 45.3.
- The biggest single-benchmark swing is EBR-Bench: 45.7% for Claude Opus 5 and 71.4% for Claude Opus 5.5.
- Claude Opus 5.5 is cheaper at $4 / $20 per million input/output tokens, against $5 / $25 for Claude Opus 5.

## Snapshot

| | Claude Opus 5 | Claude Opus 5.5 |
|---|---|---|
| Provider | Anthropic | Anthropic |
| Noometry Index | 67.8 | 68.6 |
| Rank | 4 | 3 |
| Context | 1M | 1M |
| Input $/M | $5 | $4 |
| Output $/M | $25 | $20 |
| Weights | Proprietary | Proprietary |

## Coding

- Claude Opus 5: 67.5 (#5)
- Claude Opus 5.5: 71.9 (#3)

| Benchmark | Claude Opus 5 | Claude Opus 5.5 |
|---|---|---|
| FrontierCode | 53.4% | 54.6% |
| CursorBench | 46.6% | 57.8% |
| LMArena WebDev | 1691 | 1813 |
| FrontierSWE | 52% | 62.3% |
| SciCode | 56.4% | 66.9% |
| LMArena Coding | 1534 | 1547 |
| ALE-Bench | 2,165 | 2,147 |
| DeepSWE | 73.6% | — |
| WeirdML | 91.8% | — |
| MirrorCode | — | 77.4% |

## Agentic & Tool Use

- Claude Opus 5: 55.6 (#1)
- Claude Opus 5.5: 45.3 (#15)

| Benchmark | Claude Opus 5 | Claude Opus 5.5 |
|---|---|---|
| APEX-Agents | 65.8% | 73.5% |
| GDP.pdf | 24% | 30.6% |
| Vending-Bench 2 | 11,182 | 9,235 |
| OSWorld 2.0 | 31.4% | — |
| τ²-bench Banking | 48.7% | — |
| PostTrainBench | 35% | — |
| BALROG | 63.4% | — |
| GBAEval | 79.6% | — |

## Reasoning

- Claude Opus 5: 77.2 (#4)
- Claude Opus 5.5: 80.2 (#3)

| Benchmark | Claude Opus 5 | Claude Opus 5.5 |
|---|---|---|
| ARC-AGI-2 | 90.4% | 93.3% |
| NYT Connections (extended) | 94.3% | 88.5% |
| ARC-AGI-1 | 97.5% | 98.5% |
| CritPt | 29.1% | 31.7% |
| EBR-Bench | 45.7% | 71.4% |
| LMArena Hard Prompts | 1526 | 1535 |
| Mystery Game Puzzles | 59% | 71% |
| DTBench | 97.9% | 98.9% |
| LMCA | 64.5% | 68.2% |
| Epoch Capabilities Index | 162.78 | 167.33 |
| SimpleBench | 80.6% | — |
| Chess Puzzles | 42% | — |
| Bench to the Future 3 | 0.12 | — |

## Math

- Claude Opus 5: 86.2 (#8)
- Claude Opus 5.5: 91.8 (#3)

| Benchmark | Claude Opus 5 | Claude Opus 5.5 |
|---|---|---|
| FrontierMath (Tiers 1-3) | 85.6% | 91.2% |
| FrontierMath Tier 4 | 73.2% | 95% |
| OTIS Mock AIME 2024-2025 | 98.9% | 100% |
| ProofBench | 99% | 100% |
| LMArena Math | 1531 | 1506 |
| FrontierMath Erdős | — | 2.9% |

## Knowledge

- Claude Opus 5: 66.8 (#9)
- Claude Opus 5.5: 66.4 (#10)

| Benchmark | Claude Opus 5 | Claude Opus 5.5 |
|---|---|---|
| GPQA Diamond | 93.9% | 90.6% |
| SimpleQA Verified | 59.9% | 72.2% |
| LMArena Expert | 1557 | 1547 |

## Multimodal

- Claude Opus 5: 50.8 (#8)
- Claude Opus 5.5: 57.8 (#1)

| Benchmark | Claude Opus 5 | Claude Opus 5.5 |
|---|---|---|
| LMArena Vision | 1319 | 1321 |
| Blueprint-Bench 2 | 30.4% | 51.2% |
| Furniture Assembly | 60.8% | 83.3% |
| LMArena Document | 1516 | — |

## Multilingual

- Claude Opus 5: 58.8 (#4)
- Claude Opus 5.5: 59.1 (#2)

| Benchmark | Claude Opus 5 | Claude Opus 5.5 |
|---|---|---|
| LMArena Non-English | 1501 | 1507 |
| LMArena Chinese | 1574 | 1588 |
| LMArena French | 1519 | 1514 |
| LMArena Russian | 1507 | 1520 |
| LMArena Spanish | 1519 | 1507 |
| LMArena German | 1524 | — |
| LMArena Japanese | 1516 | — |
| LMArena Korean | 1521 | — |

## Instruction Following

- Claude Opus 5: 79.2 (#7)
- Claude Opus 5.5: 80.0 (#3)

| Benchmark | Claude Opus 5 | Claude Opus 5.5 |
|---|---|---|
| LMArena Instruction Following | 1517 | 1537 |

## Long Context

- Claude Opus 5: 46.5 (#21)
- Claude Opus 5.5: 47.1 (#19)

| Benchmark | Claude Opus 5 | Claude Opus 5.5 |
|---|---|---|
| LMArena Longer Query | 1515 | 1532 |

## Writing & Preference

- Claude Opus 5: 79.2 (#1)
- Claude Opus 5.5: 78.2 (#3)

| Benchmark | Claude Opus 5 | Claude Opus 5.5 |
|---|---|---|
| LMArena Text | 1507 | 1515 |
| LMArena Creative Writing | 1491 | 1533 |
| EQ-Bench Creative Writing | 2133 | 2050 |
| LMArena Multi-Turn | 1499 | 1499 |
| EQ-Bench 4 | 1385 | — |

## FAQ

### Is Claude Opus 5 better than Claude Opus 5.5?

Claude Opus 5 and Claude Opus 5.5 score almost the same on the Noometry Index (67.8 vs 68.6), so choose on price, context window or the category you care about most.

### Which is cheaper, Claude Opus 5 or Claude Opus 5.5?

Claude Opus 5.5 is cheaper. It lists at $4 per million input tokens and $20 per million output tokens; Claude Opus 5 lists at $5 and $25.

### Is Claude Opus 5 or Claude Opus 5.5 better for coding?

Claude Opus 5.5 scores higher on coding benchmarks: 71.9 versus 67.5 in the Noometry coding category.

### Which has the bigger context window?

Both accept 1M tokens.

### How many benchmarks do Claude Opus 5 and Claude Opus 5.5 share?

42 benchmarks have published results for both models. Claude Opus 5 has 57 scored results on Noometry and Claude Opus 5.5 has 44.