# GPT-4.5 vs GPT-4 Turbo

> GPT-4.5 is the stronger model overall, scoring 37.2 to 30.5 on the Noometry Index.

- Canonical page: https://noometry.com/compare/gpt-4-5-vs-gpt-4-turbo
- Last updated: 2026-10-10
- Shared benchmarks: 25

## Summary

- They share 25 benchmarks with published results for both. GPT-4.5 scores higher in 8 categories and GPT-4 Turbo in 1 category; 9 gaps are clear of the uncertainty.
- The widest gap is in math, where GPT-4.5 leads 32.6 to 9.0.
- The biggest single-benchmark swing is MATH Level 5: 78.6% for GPT-4.5 and 46.7% for GPT-4 Turbo.

## Snapshot

| | GPT-4.5 | GPT-4 Turbo |
|---|---|---|
| Provider | OpenAI | OpenAI |
| Noometry Index | 37.2 | 30.5 |
| Rank | 208 | 292 |
| Context | — | 128K |
| Input $/M | — | $10 |
| Output $/M | — | $30 |
| Weights | Proprietary | Proprietary |

## Coding

- GPT-4.5: 42.2 (#109)
- GPT-4 Turbo: 33.8 (#249)

| Benchmark | GPT-4.5 | GPT-4 Turbo |
|---|---|---|
| WeirdML | 39.4% | 18% |
| LMArena Coding | 1396 | 1268 |
| Aider Polyglot | 44.9% | — |
| BigCodeBench Instruct | — | 48.2% |
| LiveBench Coding | 75.2% | — |
| BigCodeBench Complete | — | 58.2% |
| HumanEval+ | — | 86.6% |
| MBPP+ | — | 73.3% |

## Agentic & Tool Use

- GPT-4.5: 27.9 (#97)
- GPT-4 Turbo: —

| Benchmark | GPT-4.5 | GPT-4 Turbo |
|---|---|---|
| Cybench | 17.5% | — |
| METR Time Horizons | — | 36.7% |

## Reasoning

- GPT-4.5: 13.9 (#330)
- GPT-4 Turbo: 15.3 (#317)

| Benchmark | GPT-4.5 | GPT-4 Turbo |
|---|---|---|
| SimpleBench | 34.5% | 25.1% |
| LMArena Hard Prompts | 1403 | 1251 |
| Epoch Capabilities Index | 136.74 | 127.25 |
| ForecastBench | 61.7 | 59.4 |
| ARC-AGI-2 | 0.8% | — |
| ARC-AGI-1 | 10.3% | — |
| Chess Puzzles | — | 6% |
| EnigmaEval | 3.2% | — |
| LiveBench Reasoning | 71.1% | — |
| DTBench | — | 61.6% |
| LiveBench Data Analysis | 64.3% | — |
| LMCA | — | 9.8% |
| LiveBench | 69% | — |

## Math

- GPT-4.5: 32.6 (#211)
- GPT-4 Turbo: 9.0 (#322)

| Benchmark | GPT-4.5 | GPT-4 Turbo |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 37.8% | 6.7% |
| LMArena Math | 1412 | 1272 |
| MATH Level 5 | 78.6% | 46.7% |
| FrontierMath (Tiers 1-3) | — | 0.7% |
| LiveBench Math | 69.3% | — |

## Knowledge

- GPT-4.5: 32.5 (#211)
- GPT-4 Turbo: 24.3 (#268)

| Benchmark | GPT-4.5 | GPT-4 Turbo |
|---|---|---|
| GPQA Diamond | 68.7% | 46.6% |
| Confabulations | 13.6% | 28.4% |
| LMArena Expert | 1394 | 1223 |
| Humanity's Last Exam | 5.4% | — |
| MMLU | — | 81.3% |

## Multimodal

- GPT-4.5: 37.6 (#71)
- GPT-4 Turbo: 30.6 (#110)

| Benchmark | GPT-4.5 | GPT-4 Turbo |
|---|---|---|
| LMArena Vision | 1195 | 1090 |
| VPCT | 45% | — |

## Multilingual

- GPT-4.5: 52.5 (#83)
- GPT-4 Turbo: 40.5 (#216)

| Benchmark | GPT-4.5 | GPT-4 Turbo |
|---|---|---|
| LMArena Non-English | 1413 | 1245 |
| LMArena Chinese | 1421 | 1242 |
| LMArena French | 1418 | 1276 |
| LMArena German | 1457 | 1259 |
| LMArena Japanese | 1416 | 1194 |
| LMArena Korean | 1392 | 1187 |
| LMArena Russian | 1419 | 1259 |
| LMArena Spanish | — | 1260 |

## Instruction Following

- GPT-4.5: 72.6 (#134)
- GPT-4 Turbo: 65.8 (#216)

| Benchmark | GPT-4.5 | GPT-4 Turbo |
|---|---|---|
| LMArena Instruction Following | 1404 | 1249 |
| LiveBench Instruction Following | 72.3% | — |

## Long Context

- GPT-4.5: 40.4 (#155)
- GPT-4 Turbo: 38.0 (#206)

| Benchmark | GPT-4.5 | GPT-4 Turbo |
|---|---|---|
| LMArena Longer Query | 1406 | 1254 |
| Fiction.LiveBench | 63.9% | — |

## Writing & Preference

- GPT-4.5: 56.9 (#134)
- GPT-4 Turbo: 47.7 (#206)

| Benchmark | GPT-4.5 | GPT-4 Turbo |
|---|---|---|
| LMArena Text | 1417 | 1272 |
| LMArena Creative Writing | 1394 | 1269 |
| LMArena Multi-Turn | 1444 | 1267 |
| Short-Story Creative Writing | 75.6% | — |
| EQ-Bench Creative Writing | 1258 | — |
| LiveBench Language | 61.5% | — |

## FAQ

### Is GPT-4.5 better than GPT-4 Turbo?

GPT-4.5 is the stronger model overall, scoring 37.2 to 30.5 on the Noometry Index.

### Is GPT-4.5 or GPT-4 Turbo better for coding?

GPT-4.5 scores higher on coding benchmarks: 42.2 versus 33.8 in the Noometry coding category.

### How many benchmarks do GPT-4.5 and GPT-4 Turbo share?

25 benchmarks have published results for both models. GPT-4.5 has 42 scored results on Noometry and GPT-4 Turbo has 36.