# Claude 2 vs GPT-5 Nano

> GPT-5 Nano is the stronger model overall, scoring 33.4 to 24.7 on the Noometry Index.

- Canonical page: https://noometry.com/compare/claude-2-vs-gpt-5-nano
- Last updated: 2026-10-10
- Shared benchmarks: 5

## Summary

- They share 5 benchmarks with published results for both. Claude 2 scores higher in 1 category and GPT-5 Nano in 2 categories; 3 gaps are clear of the uncertainty.
- The widest gap is in math, where GPT-5 Nano leads 29.1 to 8.7.
- The biggest single-benchmark swing is MATH Level 5: 11.7% for Claude 2 and 95.2% for GPT-5 Nano.

## Snapshot

| | Claude 2 | GPT-5 Nano |
|---|---|---|
| Provider | Anthropic | OpenAI |
| Noometry Index | 24.7 | 33.4 |
| Rank | 342 | 234 |
| Context | — | 400K |
| Input $/M | — | $0.05 |
| Output $/M | — | $0.40 |
| Weights | Proprietary | Proprietary |

## Coding

- Claude 2: —
- GPT-5 Nano: 33.3 (#248)

| Benchmark | Claude 2 | GPT-5 Nano |
|---|---|---|
| SWE-bench Verified (bash only) | — | 34.8% |
| WeirdML | — | 38.1% |
| LMArena Coding | — | 1351 |
| ALE-Bench | — | 718.67 |
| HumanEval+ | 61.6% | — |

## Agentic & Tool Use

- Claude 2: —
- GPT-5 Nano: 25.6 (#99)

| Benchmark | Claude 2 | GPT-5 Nano |
|---|---|---|
| Terminal-Bench | — | 21.8% |
| Berkeley Function Calling Leaderboard | — | 51.5% |

## Reasoning

- Claude 2: 20.8 (#213)
- GPT-5 Nano: 16.4 (#291)

| Benchmark | Claude 2 | GPT-5 Nano |
|---|---|---|
| DTBench | 51.9% | 62.7% |
| Epoch Capabilities Index | 120.08 | 139.39 |
| ARC-AGI-2 | — | 2.6% |
| Kagi LLM Benchmark | — | 62.2% |
| ARC-AGI-1 | — | 20.7% |
| Chess Puzzles | — | 27% |
| LMArena Hard Prompts | — | 1329 |
| Mystery Game Puzzles | — | 9% |
| LMCA | — | 7.9% |
| ForecastBench | — | 59.1 |

## Math

- Claude 2: 8.7 (#314)
- GPT-5 Nano: 29.1 (#221)

| Benchmark | Claude 2 | GPT-5 Nano |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 2.5% | 81.1% |
| MATH Level 5 | 11.7% | 95.2% |
| FrontierMath (Tiers 1-3) | — | 20% |
| FrontierMath Tier 4 | — | 2.4% |
| ProofBench | — | 12% |
| Omni-MATH | — | 54.6% |
| LMArena Math | — | 1318 |
| FrontierMath (Feb 2025 set) | — | 8.3% |
| FrontierMath Tier 4 (v1) | — | 2.1% |

## Knowledge

- Claude 2: 17.0 (#282)
- GPT-5 Nano: 35.8 (#166)

| Benchmark | Claude 2 | GPT-5 Nano |
|---|---|---|
| GPQA Diamond | 34.7% | 69.4% |
| SimpleQA Verified | — | 11.7% |
| MMLU-Pro | — | 77.8% |
| Vectara Hallucination Rate | — | 10.5% |
| GPQA (HELM) | — | 67.9% |
| LMArena Expert | — | 1321 |
| MMLU | 78.5% | — |
| TriviaQA | 87.5% | — |

## Multimodal

- Claude 2: —
- GPT-5 Nano: 30.7 (#102)

| Benchmark | Claude 2 | GPT-5 Nano |
|---|---|---|
| LMArena Vision | — | 1159 |
| VPCT | — | 37.2% |

## Multilingual

- Claude 2: —
- GPT-5 Nano: 45.5 (#167)

| Benchmark | Claude 2 | GPT-5 Nano |
|---|---|---|
| LMArena Non-English | — | 1313 |
| LMArena Chinese | — | 1355 |
| LMArena German | — | 1329 |
| LMArena Korean | — | 1269 |
| LMArena Russian | — | 1296 |
| LMArena Spanish | — | 1360 |

## Instruction Following

- Claude 2: —
- GPT-5 Nano: 75.2 (#72)

| Benchmark | Claude 2 | GPT-5 Nano |
|---|---|---|
| IFEval | — | 93.2% |
| LMArena Instruction Following | — | 1306 |

## Long Context

- Claude 2: —
- GPT-5 Nano: 31.3 (#275)

| Benchmark | Claude 2 | GPT-5 Nano |
|---|---|---|
| Fiction.LiveBench | — | 44.4% |
| LMArena Longer Query | — | 1313 |

## Writing & Preference

- Claude 2: —
- GPT-5 Nano: 39.4 (#244)

| Benchmark | Claude 2 | GPT-5 Nano |
|---|---|---|
| LMArena Text | — | 1320 |
| LMArena Creative Writing | — | 1250 |
| EQ-Bench Creative Writing | — | 705 |
| WildBench | — | 80.6% |
| LMArena Multi-Turn | — | 1311 |

## FAQ

### Is Claude 2 better than GPT-5 Nano?

GPT-5 Nano is the stronger model overall, scoring 33.4 to 24.7 on the Noometry Index.

### How many benchmarks do Claude 2 and GPT-5 Nano share?

5 benchmarks have published results for both models. Claude 2 has 8 scored results on Noometry and GPT-5 Nano has 48.