# Claude Sonnet 5 vs Qwen2.5 72B Instruct

> Claude Sonnet 5 is the stronger model overall, scoring 54.5 to 31.5 on the Noometry Index. Qwen2.5 72B Instruct costs 1.6× less per token, which makes it the better buy when Claude Sonnet 5's lead doesn't matter for your workload.

- Canonical page: https://noometry.com/compare/claude-sonnet-5-vs-qwen2-5-72b-instruct
- Last updated: 2026-10-10
- Shared benchmarks: 24

## Summary

- They share 24 benchmarks with published results for both. Claude Sonnet 5 scores higher in 9 categories and Qwen2.5 72B Instruct in 0 categories; 9 gaps are clear of the uncertainty.
- The widest gap is in math, where Claude Sonnet 5 leads 66.1 to 18.1.
- The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 94.7% for Claude Sonnet 5 and 8.1% for Qwen2.5 72B Instruct.
- Qwen2.5 72B Instruct is cheaper at $1.40 / $5.60 per million input/output tokens, against $2 / $10 for Claude Sonnet 5.
- Claude Sonnet 5 accepts more context: 1M tokens versus 131K.
- Qwen2.5 72B Instruct has downloadable open weights; the other is API-only.

## Snapshot

| | Claude Sonnet 5 | Qwen2.5 72B Instruct |
|---|---|---|
| Provider | Anthropic | Alibaba (Qwen) |
| Noometry Index | 54.5 | 31.5 |
| Rank | 26 | 259 |
| Context | 1M | 131K |
| Input $/M | $2 | $1.40 |
| Output $/M | $10 | $5.60 |
| Weights | Proprietary | Open |

## Coding

- Claude Sonnet 5: 55.5 (#20)
- Qwen2.5 72B Instruct: 32.8 (#256)

| Benchmark | Claude Sonnet 5 | Qwen2.5 72B Instruct |
|---|---|---|
| WeirdML | 68.8% | 16% |
| LMArena Coding | 1483 | 1293 |
| DeepSWE | 53.8% | — |
| FrontierCode | 42.7% | — |
| CursorBench | 34.1% | — |
| LMArena WebDev | 1540 | — |
| SciCode | 53.6% | — |
| GSO | 37.3% | — |
| BigCodeBench Instruct | — | 45.8% |
| BigCodeBench Complete | — | 55.9% |
| ALE-Bench | 1,463 | — |

## Agentic & Tool Use

- Claude Sonnet 5: 42.7 (#16)
- Qwen2.5 72B Instruct: 22.0 (#124)

| Benchmark | Claude Sonnet 5 | Qwen2.5 72B Instruct |
|---|---|---|
| APEX-Agents | 54.5% | — |
| TheAgentCompany | — | 5.7% |
| BALROG | — | 16.2% |
| GBAEval | 65.3% | — |
| LMArena Search | 1194 | — |
| METR Time Horizons | — | 35.8% |
| Vending-Bench 2 | 6,378 | — |

## Reasoning

- Claude Sonnet 5: 48.2 (#37)
- Qwen2.5 72B Instruct: 21.4 (#197)

| Benchmark | Claude Sonnet 5 | Qwen2.5 72B Instruct |
|---|---|---|
| LMArena Hard Prompts | 1461 | 1271 |
| DTBench | 92.5% | 62.9% |
| LMCA | 50% | 13.4% |
| Epoch Capabilities Index | 156.34 | 129.01 |
| ForecastBench | 61.1 | 57.5 |
| SimpleBench | 60.6% | — |
| NYT Connections (extended) | 75.1% | — |
| CritPt | 16.9% | — |
| Chess Puzzles | 35% | — |
| Mystery Game Puzzles | 35% | — |
| Surface Evolver Bench | 60% | — |
| Bench to the Future 3 | 0.14 | — |
| BIG-Bench Hard | — | 79.8% |
| HellaSwag | — | 84.8% |
| PIQA | — | 82.6% |
| WinoGrande | — | 82.3% |

## Math

- Claude Sonnet 5: 66.1 (#21)
- Qwen2.5 72B Instruct: 18.1 (#281)

| Benchmark | Claude Sonnet 5 | Qwen2.5 72B Instruct |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 94.7% | 8.1% |
| LMArena Math | 1467 | 1283 |
| FrontierMath (Tiers 1-3) | 65.6% | — |
| FrontierMath Tier 4 | 29.3% | — |
| ProofBench | 77% | — |
| Omni-MATH | — | 33% |
| MATH Level 5 | — | 63.2% |

## Knowledge

- Claude Sonnet 5: 55.7 (#42)
- Qwen2.5 72B Instruct: 26.9 (#247)

| Benchmark | Claude Sonnet 5 | Qwen2.5 72B Instruct |
|---|---|---|
| GPQA Diamond | 90.5% | 49.1% |
| LMArena Expert | 1491 | 1245 |
| SimpleQA Verified | 33.7% | — |
| MMLU-Pro | — | 63.1% |
| Confabulations | — | 19.1% |
| GPQA (HELM) | — | 42.6% |
| ARC (AI2) Challenge | — | 94.5% |
| MMLU | — | 85.3% |
| TriviaQA | — | 71.9% |

## Multimodal

- Claude Sonnet 5: 42.4 (#26)
- Qwen2.5 72B Instruct: —

| Benchmark | Claude Sonnet 5 | Qwen2.5 72B Instruct |
|---|---|---|
| LMArena Vision | 1275 | — |
| Blueprint-Bench 2 | 24.9% | — |
| LMArena Document | 1466 | — |

## Multilingual

- Claude Sonnet 5: 54.0 (#55)
- Qwen2.5 72B Instruct: 41.2 (#208)

| Benchmark | Claude Sonnet 5 | Qwen2.5 72B Instruct |
|---|---|---|
| LMArena Non-English | 1431 | 1252 |
| LMArena Chinese | 1475 | 1272 |
| LMArena French | 1457 | 1279 |
| LMArena German | 1448 | 1234 |
| LMArena Japanese | 1422 | 1180 |
| LMArena Korean | 1411 | 1188 |
| LMArena Russian | 1452 | 1264 |
| LMArena Spanish | 1436 | 1255 |

## Instruction Following

- Claude Sonnet 5: 76.6 (#39)
- Qwen2.5 72B Instruct: 65.7 (#216)

| Benchmark | Claude Sonnet 5 | Qwen2.5 72B Instruct |
|---|---|---|
| LMArena Instruction Following | 1453 | 1254 |
| IFEval | — | 80.6% |

## Long Context

- Claude Sonnet 5: 44.8 (#53)
- Qwen2.5 72B Instruct: 38.9 (#183)

| Benchmark | Claude Sonnet 5 | Qwen2.5 72B Instruct |
|---|---|---|
| LMArena Longer Query | 1464 | 1282 |

## Writing & Preference

- Claude Sonnet 5: 69.4 (#24)
- Qwen2.5 72B Instruct: 46.9 (#210)

| Benchmark | Claude Sonnet 5 | Qwen2.5 72B Instruct |
|---|---|---|
| LMArena Text | 1443 | 1269 |
| LMArena Creative Writing | 1416 | 1221 |
| LMArena Multi-Turn | 1454 | 1272 |
| EQ-Bench Creative Writing | 1794 | — |
| WildBench | — | 80.2% |
| EQ-Bench 4 | 1236 | — |

## FAQ

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

Claude Sonnet 5 is the stronger model overall, scoring 54.5 to 31.5 on the Noometry Index. Qwen2.5 72B Instruct costs 1.6× less per token, which makes it the better buy when Claude Sonnet 5's lead doesn't matter for your workload.

### Which is cheaper, Claude Sonnet 5 or Qwen2.5 72B Instruct?

Qwen2.5 72B Instruct is cheaper. It lists at $1.40 per million input tokens and $5.60 per million output tokens; Claude Sonnet 5 lists at $2 and $10.

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

Claude Sonnet 5 scores higher on coding benchmarks: 55.5 versus 32.8 in the Noometry coding category.

### Which has the bigger context window?

Claude Sonnet 5 does, with 1M tokens against 131K.

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

24 benchmarks have published results for both models. Claude Sonnet 5 has 51 scored results on Noometry and Qwen2.5 72B Instruct has 43.