# GPT-4.5 vs GPT-5 Nano

> GPT-4.5 is the stronger model overall, scoring 37.2 to 33.5 on the Noometry Index.

- Canonical page: https://noometry.com/compare/gpt-4-5-vs-gpt-5-nano
- Last updated: 2026-10-12
- Shared benchmarks: 27

## Summary

- They share 27 benchmarks with published results for both. GPT-4.5 scores higher in 7 categories and GPT-5 Nano in 3 categories; 10 gaps are clear of the uncertainty.
- The widest gap is in writing & preference, where GPT-4.5 leads 56.9 to 39.1.
- The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 37.8% for GPT-4.5 and 81.1% for GPT-5 Nano.

## Snapshot

| | GPT-4.5 | GPT-5 Nano |
|---|---|---|
| Provider | OpenAI | OpenAI |
| Noometry Index | 37.2 | 33.5 |
| Rank | 208 | 241 |
| Context | — | 400K |
| Input $/M | — | $0.05 |
| Output $/M | — | $0.40 |
| Weights | Proprietary | Proprietary |

## Coding

- GPT-4.5: 42.2 (#109)
- GPT-5 Nano: 33.6 (#254)

| Benchmark | GPT-4.5 | GPT-5 Nano |
|---|---|---|
| WeirdML | 39.4% | 38.1% |
| LMArena Coding | 1396 | 1351 |
| SWE-bench Verified (bash only) | — | 34.8% |
| Aider Polyglot | 44.9% | — |
| LiveBench Coding | 75.2% | — |
| ALE-Bench | — | 718.67 |

## Agentic & Tool Use

- GPT-4.5: 27.9 (#97)
- GPT-5 Nano: 25.8 (#106)

| Benchmark | GPT-4.5 | GPT-5 Nano |
|---|---|---|
| Terminal-Bench | — | 21.8% |
| Berkeley Function Calling Leaderboard | — | 51.5% |
| Cybench | 17.5% | — |

## Reasoning

- GPT-4.5: 13.9 (#330)
- GPT-5 Nano: 16.3 (#306)

| Benchmark | GPT-4.5 | GPT-5 Nano |
|---|---|---|
| ARC-AGI-2 | 0.8% | 2.6% |
| ARC-AGI-1 | 10.3% | 20.7% |
| LMArena Hard Prompts | 1403 | 1328 |
| Epoch Capabilities Index | 136.74 | 139.38 |
| ForecastBench | 61.7 | 59.1 |
| SimpleBench | 34.5% | — |
| Kagi LLM Benchmark | — | 62.2% |
| Chess Puzzles | — | 27% |
| EnigmaEval | 3.2% | — |
| LiveBench Reasoning | 71.1% | — |
| Mystery Game Puzzles | — | 9% |
| DTBench | — | 62.7% |
| LiveBench Data Analysis | 64.3% | — |
| LMCA | — | 7.9% |
| LiveBench | 69% | — |

## Math

- GPT-4.5: 32.6 (#211)
- GPT-5 Nano: 29.4 (#241)

| Benchmark | GPT-4.5 | GPT-5 Nano |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 37.8% | 81.1% |
| LMArena Math | 1412 | 1317 |
| MATH Level 5 | 78.6% | 95.2% |
| FrontierMath (Tiers 1-3) | — | 20% |
| FrontierMath Tier 4 | — | 2.4% |
| ProofBench | — | 12% |
| Omni-MATH | — | 54.6% |
| LiveBench Math | 69.3% | — |
| FrontierMath (Feb 2025 set) | — | 8.3% |
| FrontierMath Tier 4 (v1) | — | 2.1% |

## Knowledge

- GPT-4.5: 32.5 (#211)
- GPT-5 Nano: 35.9 (#178)

| Benchmark | GPT-4.5 | GPT-5 Nano |
|---|---|---|
| GPQA Diamond | 68.7% | 69.4% |
| LMArena Expert | 1394 | 1321 |
| Humanity's Last Exam | 5.4% | — |
| SimpleQA Verified | — | 11.7% |
| MMLU-Pro | — | 77.8% |
| Confabulations | 13.6% | — |
| Vectara Hallucination Rate | — | 10.5% |
| GPQA (HELM) | — | 67.9% |

## Multimodal

- GPT-4.5: 37.6 (#71)
- GPT-5 Nano: 31.3 (#108)

| Benchmark | GPT-4.5 | GPT-5 Nano |
|---|---|---|
| LMArena Vision | 1195 | 1159 |
| VPCT | 45% | 37.2% |

## Multilingual

- GPT-4.5: 52.5 (#83)
- GPT-5 Nano: 45.3 (#172)

| Benchmark | GPT-4.5 | GPT-5 Nano |
|---|---|---|
| LMArena Non-English | 1413 | 1313 |
| LMArena Chinese | 1421 | 1356 |
| LMArena German | 1457 | 1327 |
| LMArena Japanese | 1416 | 1226 |
| LMArena Korean | 1392 | 1269 |
| LMArena Russian | 1419 | 1296 |
| LMArena French | 1418 | — |
| LMArena Spanish | — | 1360 |

## Instruction Following

- GPT-4.5: 72.6 (#134)
- GPT-5 Nano: 75.0 (#79)

| Benchmark | GPT-4.5 | GPT-5 Nano |
|---|---|---|
| LMArena Instruction Following | 1404 | 1306 |
| LiveBench Instruction Following | 72.3% | — |
| IFEval | — | 93.2% |

## Long Context

- GPT-4.5: 40.4 (#155)
- GPT-5 Nano: 31.3 (#281)

| Benchmark | GPT-4.5 | GPT-5 Nano |
|---|---|---|
| Fiction.LiveBench | 63.9% | 44.4% |
| LMArena Longer Query | 1406 | 1312 |

## Writing & Preference

- GPT-4.5: 56.9 (#134)
- GPT-5 Nano: 39.1 (#249)

| Benchmark | GPT-4.5 | GPT-5 Nano |
|---|---|---|
| LMArena Text | 1417 | 1320 |
| LMArena Creative Writing | 1394 | 1249 |
| EQ-Bench Creative Writing | 1258 | 705 |
| LMArena Multi-Turn | 1444 | 1311 |
| Short-Story Creative Writing | 75.6% | — |
| WildBench | — | 80.6% |
| LiveBench Language | 61.5% | — |

## FAQ

### Is GPT-4.5 better than GPT-5 Nano?

GPT-4.5 is the stronger model overall, scoring 37.2 to 33.5 on the Noometry Index.

### Is GPT-4.5 or GPT-5 Nano better for coding?

GPT-4.5 scores higher on coding benchmarks: 42.2 versus 33.6 in the Noometry coding category.

### How many benchmarks do GPT-4.5 and GPT-5 Nano share?

27 benchmarks have published results for both models. GPT-4.5 has 42 scored results on Noometry and GPT-5 Nano has 49.