# GPT-4.5 vs GPT-5.1

> GPT-5.1 is the stronger model overall, scoring 49.0 to 37.2 on the Noometry Index.

- Canonical page: https://noometry.com/compare/gpt-4-5-vs-gpt-5-1
- Last updated: 2026-10-10
- Shared benchmarks: 35

## Summary

- They share 35 benchmarks with published results for both. GPT-4.5 scores higher in 0 categories and GPT-5.1 in 10 categories; 10 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where GPT-5.1 leads 39.8 to 13.9.
- The biggest single-benchmark swing is ARC-AGI-1: 10.3% for GPT-4.5 and 72.8% for GPT-5.1.

## Snapshot

| | GPT-4.5 | GPT-5.1 |
|---|---|---|
| Provider | OpenAI | OpenAI |
| Noometry Index | 37.2 | 49.0 |
| Rank | 208 | 53 |
| Context | — | 400K |
| Input $/M | — | $1.25 |
| Output $/M | — | $10 |
| Weights | Proprietary | Proprietary |

## Coding

- GPT-4.5: 42.2 (#109)
- GPT-5.1: 46.4 (#66)

| Benchmark | GPT-4.5 | GPT-5.1 |
|---|---|---|
| WeirdML | 39.4% | 60.8% |
| LiveBench Coding | 75.2% | 72.5% |
| LMArena Coding | 1396 | 1454 |
| SWE-bench Verified | — | 68% |
| SWE-bench Verified (bash only) | — | 66% |
| Aider Polyglot | 44.9% | — |
| LMArena WebDev | — | 1395 |
| SciCode | — | 43.3% |
| GSO | — | 13.7% |
| ALE-Bench | — | 1,192 |

## Agentic & Tool Use

- GPT-4.5: 27.9 (#97)
- GPT-5.1: 32.7 (#60)

| Benchmark | GPT-4.5 | GPT-5.1 |
|---|---|---|
| Terminal-Bench | — | 47.6% |
| Cybench | 17.5% | — |
| DeepResearch Bench | — | 42.8% |
| LMArena Search | — | 1199 |
| Vending-Bench 2 | — | 1,473 |

## Reasoning

- GPT-4.5: 13.9 (#330)
- GPT-5.1: 39.8 (#58)

| Benchmark | GPT-4.5 | GPT-5.1 |
|---|---|---|
| ARC-AGI-2 | 0.8% | 17.6% |
| SimpleBench | 34.5% | 53.2% |
| ARC-AGI-1 | 10.3% | 72.8% |
| EnigmaEval | 3.2% | 11.2% |
| LiveBench Reasoning | 71.1% | 95.8% |
| LMArena Hard Prompts | 1403 | 1457 |
| LiveBench Data Analysis | 64.3% | 72.1% |
| Epoch Capabilities Index | 136.74 | 149.64 |
| ForecastBench | 61.7 | 58.1 |
| LiveBench | 69% | 78.8% |
| CritPt | — | 4.9% |
| Chess Puzzles | — | 32% |
| Mystery Game Puzzles | — | 19% |
| DTBench | — | 90.1% |
| LMCA | — | 43.9% |

## Math

- GPT-4.5: 32.6 (#211)
- GPT-5.1: 52.2 (#51)

| Benchmark | GPT-4.5 | GPT-5.1 |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 37.8% | 88.6% |
| LiveBench Math | 69.3% | 94.5% |
| LMArena Math | 1412 | 1447 |
| Omni-MATH | — | 46.4% |
| MATH Level 5 | 78.6% | — |
| FrontierMath (Feb 2025 set) | — | 31% |
| FrontierMath Tier 4 (v1) | — | 12.5% |

## Knowledge

- GPT-4.5: 32.5 (#211)
- GPT-5.1: 50.6 (#71)

| Benchmark | GPT-4.5 | GPT-5.1 |
|---|---|---|
| GPQA Diamond | 68.7% | 87.6% |
| Humanity's Last Exam | 5.4% | 23.7% |
| LMArena Expert | 1394 | 1470 |
| SimpleQA Verified | — | 48% |
| MMLU-Pro | — | 57.9% |
| Confabulations | 13.6% | — |
| Vectara Hallucination Rate | — | 10.9% |
| GPQA (HELM) | — | 44.2% |

## Multimodal

- GPT-4.5: 37.6 (#71)
- GPT-5.1: 44.8 (#19)

| Benchmark | GPT-4.5 | GPT-5.1 |
|---|---|---|
| LMArena Vision | 1195 | 1250 |
| VPCT | 45% | 58.7% |
| LMArena Document | — | 1403 |

## Multilingual

- GPT-4.5: 52.5 (#83)
- GPT-5.1: 53.8 (#56)

| Benchmark | GPT-4.5 | GPT-5.1 |
|---|---|---|
| LMArena Non-English | 1413 | 1431 |
| LMArena Chinese | 1421 | 1495 |
| LMArena French | 1418 | 1450 |
| LMArena German | 1457 | 1438 |
| LMArena Japanese | 1416 | 1453 |
| LMArena Korean | 1392 | 1401 |
| LMArena Russian | 1419 | 1435 |
| LMArena Spanish | — | 1433 |

## Instruction Following

- GPT-4.5: 72.6 (#134)
- GPT-5.1: 83.9 (#1)

| Benchmark | GPT-4.5 | GPT-5.1 |
|---|---|---|
| LiveBench Instruction Following | 72.3% | 93.3% |
| LMArena Instruction Following | 1404 | 1443 |
| IFEval | — | 93.5% |

## Long Context

- GPT-4.5: 40.4 (#155)
- GPT-5.1: 47.6 (#14)

| Benchmark | GPT-4.5 | GPT-5.1 |
|---|---|---|
| LMArena Longer Query | 1406 | 1447 |
| Fiction.LiveBench | 63.9% | — |
| CL-bench | — | 23.7% |
| CL-bench Life | — | 17.3% |

## Writing & Preference

- GPT-4.5: 56.9 (#134)
- GPT-5.1: 64.5 (#55)

| Benchmark | GPT-4.5 | GPT-5.1 |
|---|---|---|
| LMArena Text | 1417 | 1443 |
| LMArena Creative Writing | 1394 | 1427 |
| LMArena Multi-Turn | 1444 | 1450 |
| LiveBench Language | 61.5% | 80.2% |
| Short-Story Creative Writing | 75.6% | — |
| EQ-Bench Creative Writing | 1258 | — |
| WildBench | — | 86.3% |

## FAQ

### Is GPT-4.5 better than GPT-5.1?

GPT-5.1 is the stronger model overall, scoring 49.0 to 37.2 on the Noometry Index.

### Is GPT-4.5 or GPT-5.1 better for coding?

GPT-5.1 scores higher on coding benchmarks: 46.4 versus 42.2 in the Noometry coding category.

### How many benchmarks do GPT-4.5 and GPT-5.1 share?

35 benchmarks have published results for both models. GPT-4.5 has 42 scored results on Noometry and GPT-5.1 has 63.