# Claude Opus 5.5 vs Claude Sonnet 5

> Claude Opus 5.5 is the stronger model overall, scoring 68.6 to 54.6 on the Noometry Index. Claude Sonnet 5 costs 2.0× less per token, which makes it the better buy when Claude Opus 5.5's lead doesn't matter for your workload.

- Canonical page: https://noometry.com/compare/claude-opus-5-5-vs-claude-sonnet-5
- Last updated: 2026-10-12
- Shared benchmarks: 36

## Summary

- They share 36 benchmarks with published results for both. Claude Opus 5.5 scores higher in 10 categories and Claude Sonnet 5 in 0 categories; 10 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where Claude Opus 5.5 leads 80.2 to 49.1.
- The biggest single-benchmark swing is FrontierMath Tier 4: 95% for Claude Opus 5.5 and 29.3% for Claude Sonnet 5.
- Claude Sonnet 5 is cheaper at $2 / $10 per million input/output tokens, against $4 / $20 for Claude Opus 5.5.

## Snapshot

| | Claude Opus 5.5 | Claude Sonnet 5 |
|---|---|---|
| Provider | Anthropic | Anthropic |
| Noometry Index | 68.6 | 54.6 |
| Rank | 3 | 29 |
| Context | 1M | 1M |
| Input $/M | $4 | $2 |
| Output $/M | $20 | $10 |
| Weights | Proprietary | Proprietary |

## Coding

- Claude Opus 5.5: 71.9 (#3)
- Claude Sonnet 5: 55.5 (#26)

| Benchmark | Claude Opus 5.5 | Claude Sonnet 5 |
|---|---|---|
| FrontierCode | 54.6% | 42.7% |
| CursorBench | 57.8% | 34.1% |
| LMArena WebDev | 1813 | 1541 |
| SciCode | 66.9% | 54.3% |
| LMArena Coding | 1547 | 1483 |
| ALE-Bench | 2,147 | 1,463 |
| DeepSWE | — | 53.8% |
| FrontierSWE | 62.3% | — |
| GSO | — | 37.3% |
| WeirdML | — | 68.8% |
| MirrorCode | 77.4% | — |

## Agentic & Tool Use

- Claude Opus 5.5: 45.3 (#15)
- Claude Sonnet 5: 42.8 (#18)

| Benchmark | Claude Opus 5.5 | Claude Sonnet 5 |
|---|---|---|
| APEX-Agents | 73.5% | 54.5% |
| Vending-Bench 2 | 9,235 | 6,378 |
| GBAEval | — | 65.3% |
| GDP.pdf | 30.6% | — |
| LMArena Search | — | 1194 |

## Reasoning

- Claude Opus 5.5: 80.2 (#3)
- Claude Sonnet 5: 49.1 (#39)

| Benchmark | Claude Opus 5.5 | Claude Sonnet 5 |
|---|---|---|
| NYT Connections (extended) | 88.5% | 75.1% |
| CritPt | 31.7% | 16.9% |
| LMArena Hard Prompts | 1535 | 1461 |
| Mystery Game Puzzles | 71% | 35% |
| DTBench | 98.9% | 92.5% |
| LMCA | 68.2% | 50% |
| Epoch Capabilities Index | 167.33 | 156.21 |
| ARC-AGI-2 | 93.3% | — |
| SimpleBench | — | 60.6% |
| ARC-AGI-1 | 98.5% | — |
| Chess Puzzles | — | 35% |
| EBR-Bench | 71.4% | — |
| Surface Evolver Bench | — | 60% |
| Bench to the Future 3 | — | 0.14 |
| ForecastBench | — | 61.1 |

## Math

- Claude Opus 5.5: 91.8 (#3)
- Claude Sonnet 5: 66.2 (#27)

| Benchmark | Claude Opus 5.5 | Claude Sonnet 5 |
|---|---|---|
| FrontierMath (Tiers 1-3) | 91.2% | 65.6% |
| FrontierMath Tier 4 | 95% | 29.3% |
| OTIS Mock AIME 2024-2025 | 100% | 94.7% |
| ProofBench | 100% | 77% |
| LMArena Math | 1506 | 1467 |
| FrontierMath Erdős | 2.9% | — |

## Knowledge

- Claude Opus 5.5: 66.4 (#10)
- Claude Sonnet 5: 55.6 (#47)

| Benchmark | Claude Opus 5.5 | Claude Sonnet 5 |
|---|---|---|
| GPQA Diamond | 90.6% | 90.5% |
| SimpleQA Verified | 72.2% | 33.7% |
| LMArena Expert | 1547 | 1490 |

## Multimodal

- Claude Opus 5.5: 57.8 (#1)
- Claude Sonnet 5: 42.4 (#31)

| Benchmark | Claude Opus 5.5 | Claude Sonnet 5 |
|---|---|---|
| LMArena Vision | 1321 | 1274 |
| Blueprint-Bench 2 | 51.2% | 24.9% |
| Furniture Assembly | 83.3% | — |
| LMArena Document | — | 1466 |

## Multilingual

- Claude Opus 5.5: 59.1 (#2)
- Claude Sonnet 5: 53.8 (#55)

| Benchmark | Claude Opus 5.5 | Claude Sonnet 5 |
|---|---|---|
| LMArena Non-English | 1507 | 1431 |
| LMArena Chinese | 1588 | 1477 |
| LMArena French | 1514 | 1460 |
| LMArena Russian | 1520 | 1451 |
| LMArena Spanish | 1507 | 1437 |
| LMArena German | — | 1440 |
| LMArena Japanese | — | 1422 |
| LMArena Korean | — | 1411 |

## Instruction Following

- Claude Opus 5.5: 80.0 (#3)
- Claude Sonnet 5: 76.3 (#41)

| Benchmark | Claude Opus 5.5 | Claude Sonnet 5 |
|---|---|---|
| LMArena Instruction Following | 1537 | 1452 |

## Long Context

- Claude Opus 5.5: 47.1 (#19)
- Claude Sonnet 5: 44.8 (#55)

| Benchmark | Claude Opus 5.5 | Claude Sonnet 5 |
|---|---|---|
| LMArena Longer Query | 1532 | 1463 |

## Writing & Preference

- Claude Opus 5.5: 78.2 (#3)
- Claude Sonnet 5: 69.2 (#25)

| Benchmark | Claude Opus 5.5 | Claude Sonnet 5 |
|---|---|---|
| LMArena Text | 1515 | 1442 |
| LMArena Creative Writing | 1533 | 1416 |
| EQ-Bench Creative Writing | 2050 | 1794 |
| LMArena Multi-Turn | 1499 | 1454 |
| EQ-Bench 4 | — | 1236 |

## FAQ

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

Claude Opus 5.5 is the stronger model overall, scoring 68.6 to 54.6 on the Noometry Index. Claude Sonnet 5 costs 2.0× less per token, which makes it the better buy when Claude Opus 5.5's lead doesn't matter for your workload.

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

Claude Sonnet 5 is cheaper. It lists at $2 per million input tokens and $10 per million output tokens; Claude Opus 5.5 lists at $4 and $20.

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

Claude Opus 5.5 scores higher on coding benchmarks: 71.9 versus 55.5 in the Noometry coding category.

### Which has the bigger context window?

Both accept 1M tokens.

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

36 benchmarks have published results for both models. Claude Opus 5.5 has 44 scored results on Noometry and Claude Sonnet 5 has 51.