# Claude Opus 5.5 vs GPT-5

> Claude Opus 5.5 is the stronger model overall, scoring 68.5 to 50.9 on the Noometry Index. GPT-5 costs 2.3× 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-gpt-5
- Last updated: 2026-10-10
- Shared benchmarks: 27

## Summary

- They share 27 benchmarks with published results for both. Claude Opus 5.5 scores higher in 9 categories and GPT-5 in 1 category; 10 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where Claude Opus 5.5 leads 79.5 to 38.3.
- The biggest single-benchmark swing is ARC-AGI-2: 92.5% for Claude Opus 5.5 and 9.9% for GPT-5.
- GPT-5 is cheaper at $1.25 / $10 per million input/output tokens, against $4 / $20 for Claude Opus 5.5.
- Claude Opus 5.5 accepts more context: 1M tokens versus 400K.

## Snapshot

| | Claude Opus 5.5 | GPT-5 |
|---|---|---|
| Provider | Anthropic | OpenAI |
| Noometry Index | 68.5 | 50.9 |
| Rank | 3 | 37 |
| Context | 1M | 400K |
| Input $/M | $4 | $1.25 |
| Output $/M | $20 | $10 |
| Weights | Proprietary | Proprietary |

## Coding

- Claude Opus 5.5: 71.4 (#1)
- GPT-5: 50.7 (#39)

| Benchmark | Claude Opus 5.5 | GPT-5 |
|---|---|---|
| LMArena WebDev | 1820 | 1417 |
| SciCode | 66.9% | 42.9% |
| LMArena Coding | 1545 | 1436 |
| SWE-bench Verified | — | 73.6% |
| FrontierCode | 54.4% | — |
| SWE-bench Verified (bash only) | — | 65% |
| Aider Polyglot | — | 88% |
| CursorBench | 57.8% | — |
| FrontierSWE | 62.3% | — |
| GSO | — | 6.9% |
| WeirdML | — | 60.7% |
| MirrorCode | 77.4% | — |
| ALE-Bench | — | 1,162 |
| AlgoTune | — | 1.67 |

## Agentic & Tool Use

- Claude Opus 5.5: 45.3 (#13)
- GPT-5: 33.0 (#53)

| Benchmark | Claude Opus 5.5 | GPT-5 |
|---|---|---|
| Terminal-Bench | — | 49.6% |
| APEX-Agents | 73.5% | — |
| GDPval | — | 34.8% |
| Remote Labor Index | — | 1.7% |
| DeepResearch Bench | — | 49.6% |
| BALROG | — | 32.8% |
| GDP.pdf | 30.6% | — |
| LMArena Search | — | 1133 |
| METR Time Horizons | — | 69.6% |

## Reasoning

- Claude Opus 5.5: 79.5 (#2)
- GPT-5: 38.3 (#55)

| Benchmark | Claude Opus 5.5 | GPT-5 |
|---|---|---|
| ARC-AGI-2 | 92.5% | 9.9% |
| ARC-AGI-1 | 97.5% | 65.7% |
| CritPt | 31.7% | 12.6% |
| EBR-Bench | 71.4% | 12.7% |
| LMArena Hard Prompts | 1545 | 1415 |
| Mystery Game Puzzles | 71% | 23% |
| DTBench | 98.9% | 90.7% |
| LMCA | 68.2% | 40% |
| SimpleBench | — | 56.7% |
| Kagi LLM Benchmark | — | 72.7% |
| NYT Connections (extended) | 88.5% | — |
| Chess Puzzles | — | 37% |
| EnigmaEval | — | 10.5% |
| Epoch Capabilities Index | — | 150 |
| ForecastBench | — | 61.4 |

## Math

- Claude Opus 5.5: 92.4 (#2)
- GPT-5: 54.6 (#36)

| Benchmark | Claude Opus 5.5 | GPT-5 |
|---|---|---|
| FrontierMath (Tiers 1-3) | 91.2% | 55.4% |
| FrontierMath Tier 4 | 95% | 22% |
| OTIS Mock AIME 2024-2025 | 100% | 91.4% |
| ProofBench | 100% | 18% |
| Omni-MATH | — | 64.7% |
| LMArena Math | — | 1406 |
| MATH Level 5 | — | 98.1% |
| FrontierMath (Feb 2025 set) | — | 32.4% |
| FrontierMath Tier 4 (v1) | — | 12.5% |

## Knowledge

- Claude Opus 5.5: 66.6 (#9)
- GPT-5: 56.5 (#37)

| Benchmark | Claude Opus 5.5 | GPT-5 |
|---|---|---|
| GPQA Diamond | 90.6% | 86.2% |
| SimpleQA Verified | 72.2% | 50.1% |
| LMArena Expert | 1564 | 1420 |
| Humanity's Last Exam | — | 25.3% |
| MMLU-Pro | — | 86.3% |
| Confabulations | — | 10.3% |
| Vectara Hallucination Rate | — | 14.7% |
| GPQA (HELM) | — | 79.2% |

## Multimodal

- Claude Opus 5.5: 56.5 (#2)
- GPT-5: 46.1 (#12)

| Benchmark | Claude Opus 5.5 | GPT-5 |
|---|---|---|
| LMArena Vision | — | 1232 |
| GeoBench | — | 81% |
| VPCT | — | 66% |
| Furniture Assembly | 83.3% | — |

## Multilingual

- Claude Opus 5.5: 59.8 (#1)
- GPT-5: 51.5 (#105)

| Benchmark | Claude Opus 5.5 | GPT-5 |
|---|---|---|
| LMArena Non-English | 1513 | 1397 |
| LMArena Chinese | 1572 | 1423 |
| LMArena Russian | 1532 | 1406 |
| LMArena French | — | 1409 |
| LMArena German | — | 1415 |
| LMArena Japanese | — | 1411 |
| LMArena Korean | — | 1361 |
| LMArena Spanish | — | 1398 |

## Instruction Following

- Claude Opus 5.5: 80.1 (#2)
- GPT-5: 74.0 (#107)

| Benchmark | Claude Opus 5.5 | GPT-5 |
|---|---|---|
| LMArena Instruction Following | 1535 | 1388 |
| IFEval | — | 87.5% |

## Long Context

- Claude Opus 5.5: 47.3 (#18)
- GPT-5: 69.5 (#2)

| Benchmark | Claude Opus 5.5 | GPT-5 |
|---|---|---|
| LMArena Longer Query | 1540 | 1399 |
| Fiction.LiveBench | — | 97.2% |

## Writing & Preference

- Claude Opus 5.5: 78.7 (#3)
- GPT-5: 63.5 (#63)

| Benchmark | Claude Opus 5.5 | GPT-5 |
|---|---|---|
| LMArena Text | 1518 | 1406 |
| LMArena Creative Writing | 1538 | 1365 |
| EQ-Bench Creative Writing | 2050 | 1627 |
| LMArena Multi-Turn | 1520 | 1426 |
| Short-Story Creative Writing | — | 86% |
| WildBench | — | 85.7% |

## FAQ

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

Claude Opus 5.5 is the stronger model overall, scoring 68.5 to 50.9 on the Noometry Index. GPT-5 costs 2.3× 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 GPT-5?

GPT-5 is cheaper. It lists at $1.25 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 GPT-5 better for coding?

Claude Opus 5.5 scores higher on coding benchmarks: 71.4 versus 50.7 in the Noometry coding category.

### Which has the bigger context window?

Claude Opus 5.5 does, with 1M tokens against 400K.

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

27 benchmarks have published results for both models. Claude Opus 5.5 has 35 scored results on Noometry and GPT-5 has 69.