# GPT-4.5 vs Qwen2.5 72B Instruct

> GPT-4.5 is the stronger model overall, scoring 36.9 to 31.5 on the Noometry Index.

- Canonical page: https://noometry.com/compare/gpt-4-5-vs-qwen2-5-72b-instruct
- Last updated: 2026-10-10
- Shared benchmarks: 23

## Summary

- They share 23 benchmarks with published results for both. GPT-4.5 scores higher in 8 categories and Qwen2.5 72B Instruct in 1 category; 9 gaps are clear of the uncertainty.
- The widest gap is in math, where GPT-4.5 leads 31.0 to 18.1.
- The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 37.8% for GPT-4.5 and 8.1% for Qwen2.5 72B Instruct.
- Qwen2.5 72B Instruct has downloadable open weights; the other is API-only.

## Snapshot

| | GPT-4.5 | Qwen2.5 72B Instruct |
|---|---|---|
| Provider | OpenAI | Alibaba (Qwen) |
| Noometry Index | 36.9 | 31.5 |
| Rank | 200 | 259 |
| Context | — | 131K |
| Input $/M | — | $1.40 |
| Output $/M | — | $5.60 |
| Weights | Proprietary | Open |

## Coding

- GPT-4.5: 41.8 (#102)
- Qwen2.5 72B Instruct: 32.8 (#256)

| Benchmark | GPT-4.5 | Qwen2.5 72B Instruct |
|---|---|---|
| WeirdML | 39.4% | 16% |
| LMArena Coding | 1397 | 1293 |
| Aider Polyglot | 44.9% | — |
| BigCodeBench Instruct | — | 45.8% |
| LiveBench Coding | 75.2% | — |
| BigCodeBench Complete | — | 55.9% |

## Agentic & Tool Use

- GPT-4.5: 27.6 (#92)
- Qwen2.5 72B Instruct: 22.0 (#124)

| Benchmark | GPT-4.5 | Qwen2.5 72B Instruct |
|---|---|---|
| TheAgentCompany | — | 5.7% |
| Cybench | 17.5% | — |
| BALROG | — | 16.2% |
| METR Time Horizons | — | 35.8% |

## Reasoning

- GPT-4.5: 14.1 (#320)
- Qwen2.5 72B Instruct: 21.4 (#197)

| Benchmark | GPT-4.5 | Qwen2.5 72B Instruct |
|---|---|---|
| LMArena Hard Prompts | 1404 | 1271 |
| Epoch Capabilities Index | 136.74 | 129.01 |
| ForecastBench | 61.7 | 57.5 |
| ARC-AGI-2 | 0.8% | — |
| SimpleBench | 34.5% | — |
| ARC-AGI-1 | 10.3% | — |
| EnigmaEval | 3.2% | — |
| LiveBench Reasoning | 71.1% | — |
| DTBench | — | 62.9% |
| LiveBench Data Analysis | 64.3% | — |
| LMCA | — | 13.4% |
| BIG-Bench Hard | — | 79.8% |
| HellaSwag | — | 84.8% |
| LiveBench | 69% | — |
| PIQA | — | 82.6% |
| WinoGrande | — | 82.3% |

## Math

- GPT-4.5: 31.0 (#200)
- Qwen2.5 72B Instruct: 18.1 (#281)

| Benchmark | GPT-4.5 | Qwen2.5 72B Instruct |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 37.8% | 8.1% |
| LMArena Math | 1412 | 1283 |
| MATH Level 5 | 78.6% | 63.2% |
| Omni-MATH | — | 33% |
| LiveBench Math | 69.3% | — |

## Knowledge

- GPT-4.5: 32.4 (#205)
- Qwen2.5 72B Instruct: 26.9 (#247)

| Benchmark | GPT-4.5 | Qwen2.5 72B Instruct |
|---|---|---|
| GPQA Diamond | 68.7% | 49.1% |
| Confabulations | 13.6% | 19.1% |
| LMArena Expert | 1394 | 1245 |
| Humanity's Last Exam | 5.4% | — |
| MMLU-Pro | — | 63.1% |
| GPQA (HELM) | — | 42.6% |
| ARC (AI2) Challenge | — | 94.5% |
| MMLU | — | 85.3% |
| TriviaQA | — | 71.9% |

## Multimodal

- GPT-4.5: 37.0 (#68)
- Qwen2.5 72B Instruct: —

| Benchmark | GPT-4.5 | Qwen2.5 72B Instruct |
|---|---|---|
| LMArena Vision | 1195 | — |
| VPCT | 45% | — |

## Multilingual

- GPT-4.5: 52.7 (#78)
- Qwen2.5 72B Instruct: 41.2 (#208)

| Benchmark | GPT-4.5 | Qwen2.5 72B Instruct |
|---|---|---|
| LMArena Non-English | 1413 | 1252 |
| LMArena Chinese | 1421 | 1272 |
| LMArena French | 1417 | 1279 |
| LMArena German | 1457 | 1234 |
| LMArena Japanese | 1416 | 1180 |
| LMArena Korean | 1392 | 1188 |
| LMArena Russian | 1419 | 1264 |
| LMArena Spanish | — | 1255 |

## Instruction Following

- GPT-4.5: 72.8 (#129)
- Qwen2.5 72B Instruct: 65.7 (#216)

| Benchmark | GPT-4.5 | Qwen2.5 72B Instruct |
|---|---|---|
| LMArena Instruction Following | 1404 | 1254 |
| LiveBench Instruction Following | 72.3% | — |
| IFEval | — | 80.6% |

## Long Context

- GPT-4.5: 40.4 (#150)
- Qwen2.5 72B Instruct: 38.9 (#183)

| Benchmark | GPT-4.5 | Qwen2.5 72B Instruct |
|---|---|---|
| LMArena Longer Query | 1406 | 1282 |
| Fiction.LiveBench | 63.9% | — |

## Writing & Preference

- GPT-4.5: 57.1 (#129)
- Qwen2.5 72B Instruct: 46.9 (#210)

| Benchmark | GPT-4.5 | Qwen2.5 72B Instruct |
|---|---|---|
| LMArena Text | 1417 | 1269 |
| LMArena Creative Writing | 1394 | 1221 |
| LMArena Multi-Turn | 1444 | 1272 |
| Short-Story Creative Writing | 75.6% | — |
| EQ-Bench Creative Writing | 1258 | — |
| WildBench | — | 80.2% |
| LiveBench Language | 61.5% | — |

## FAQ

### Is GPT-4.5 better than Qwen2.5 72B Instruct?

GPT-4.5 is the stronger model overall, scoring 36.9 to 31.5 on the Noometry Index.

### Is GPT-4.5 or Qwen2.5 72B Instruct better for coding?

GPT-4.5 scores higher on coding benchmarks: 41.8 versus 32.8 in the Noometry coding category.

### How many benchmarks do GPT-4.5 and Qwen2.5 72B Instruct share?

23 benchmarks have published results for both models. GPT-4.5 has 42 scored results on Noometry and Qwen2.5 72B Instruct has 43.