# Claude 2.1 vs Qwen2.5 72B Instruct

> Qwen2.5 72B Instruct is the stronger model overall, scoring 31.9 to 25.2 on the Noometry Index.

- Canonical page: https://noometry.com/compare/claude-2-1-vs-qwen2-5-72b-instruct
- Last updated: 2026-10-12
- Shared benchmarks: 7

## Summary

- They share 7 benchmarks with published results for both. Claude 2.1 scores higher in 0 categories and Qwen2.5 72B Instruct in 4 categories; 3 gaps are clear of the uncertainty.
- The widest gap is in knowledge, where Qwen2.5 72B Instruct leads 27.0 to 15.4.
- The biggest single-benchmark swing is GPQA Diamond: 33% for Claude 2.1 and 49.1% for Qwen2.5 72B Instruct.
- Qwen2.5 72B Instruct has downloadable open weights; the other is API-only.

## Snapshot

| | Claude 2.1 | Qwen2.5 72B Instruct |
|---|---|---|
| Provider | Anthropic | Alibaba (Qwen) |
| Noometry Index | 25.2 | 31.9 |
| Rank | 345 | 267 |
| Context | — | 131K |
| Input $/M | — | $1.40 |
| Output $/M | — | $5.60 |
| Weights | Proprietary | Open |

## Coding

- Claude 2.1: 26.2 (#327)
- Qwen2.5 72B Instruct: 33.2 (#260)

| Benchmark | Claude 2.1 | Qwen2.5 72B Instruct |
|---|---|---|
| WeirdML | 7.1% | 16% |
| BigCodeBench Instruct | — | 45.8% |
| LMArena Coding | — | 1292 |
| BigCodeBench Complete | — | 55.9% |

## Agentic & Tool Use

- Claude 2.1: —
- Qwen2.5 72B Instruct: 22.1 (#133)

| Benchmark | Claude 2.1 | Qwen2.5 72B Instruct |
|---|---|---|
| TheAgentCompany | — | 5.7% |
| BALROG | — | 16.2% |
| METR Time Horizons | — | 35.8% |

## Reasoning

- Claude 2.1: 21.4 (#221)
- Qwen2.5 72B Instruct: 22.3 (#199)

| Benchmark | Claude 2.1 | Qwen2.5 72B Instruct |
|---|---|---|
| DTBench | 51% | 62.9% |
| Epoch Capabilities Index | 119.27 | 129 |
| ForecastBench | 54.2 | 57.5 |
| LMArena Hard Prompts | — | 1271 |
| LMCA | — | 13.4% |
| BIG-Bench Hard | — | 79.8% |
| HellaSwag | — | 84.8% |
| PIQA | — | 82.6% |
| WinoGrande | — | 82.3% |

## Math

- Claude 2.1: 10.2 (#315)
- Qwen2.5 72B Instruct: 19.3 (#287)

| Benchmark | Claude 2.1 | Qwen2.5 72B Instruct |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 1.9% | 8.1% |
| Omni-MATH | — | 33% |
| LMArena Math | — | 1283 |
| MATH Level 5 | — | 63.2% |

## Knowledge

- Claude 2.1: 15.4 (#292)
- Qwen2.5 72B Instruct: 27.0 (#253)

| Benchmark | Claude 2.1 | Qwen2.5 72B Instruct |
|---|---|---|
| GPQA Diamond | 33% | 49.1% |
| MMLU | 73.5% | 85.3% |
| MMLU-Pro | — | 63.1% |
| Confabulations | — | 19.1% |
| GPQA (HELM) | — | 42.6% |
| LMArena Expert | — | 1245 |
| ARC (AI2) Challenge | — | 94.5% |
| TriviaQA | — | 71.9% |

## Multilingual

- Claude 2.1: —
- Qwen2.5 72B Instruct: 41.0 (#213)

| Benchmark | Claude 2.1 | Qwen2.5 72B Instruct |
|---|---|---|
| LMArena Non-English | — | 1252 |
| LMArena Chinese | — | 1272 |
| LMArena French | — | 1280 |
| LMArena German | — | 1234 |
| LMArena Japanese | — | 1180 |
| LMArena Korean | — | 1188 |
| LMArena Russian | — | 1264 |
| LMArena Spanish | — | 1256 |

## Instruction Following

- Claude 2.1: —
- Qwen2.5 72B Instruct: 65.5 (#221)

| Benchmark | Claude 2.1 | Qwen2.5 72B Instruct |
|---|---|---|
| IFEval | — | 80.6% |
| LMArena Instruction Following | — | 1254 |

## Long Context

- Claude 2.1: —
- Qwen2.5 72B Instruct: 38.9 (#188)

| Benchmark | Claude 2.1 | Qwen2.5 72B Instruct |
|---|---|---|
| LMArena Longer Query | — | 1282 |

## Writing & Preference

- Claude 2.1: —
- Qwen2.5 72B Instruct: 46.7 (#215)

| Benchmark | Claude 2.1 | Qwen2.5 72B Instruct |
|---|---|---|
| LMArena Text | — | 1269 |
| LMArena Creative Writing | — | 1221 |
| WildBench | — | 80.2% |
| LMArena Multi-Turn | — | 1272 |

## FAQ

### Is Claude 2.1 better than Qwen2.5 72B Instruct?

Qwen2.5 72B Instruct is the stronger model overall, scoring 31.9 to 25.2 on the Noometry Index.

### Is Claude 2.1 or Qwen2.5 72B Instruct better for coding?

Qwen2.5 72B Instruct scores higher on coding benchmarks: 33.2 versus 26.2 in the Noometry coding category.

### How many benchmarks do Claude 2.1 and Qwen2.5 72B Instruct share?

7 benchmarks have published results for both models. Claude 2.1 has 7 scored results on Noometry and Qwen2.5 72B Instruct has 43.