# Claude Opus 4 vs Qwen2.5 72B Instruct

> Claude Opus 4 is the stronger model overall, scoring 42.9 to 31.5 on the Noometry Index. Qwen2.5 72B Instruct costs 12× less per token, which makes it the better buy when Claude Opus 4's lead doesn't matter for your workload.

- Canonical page: https://noometry.com/compare/claude-opus-4-vs-qwen2-5-72b-instruct
- Last updated: 2026-10-10
- Shared benchmarks: 32

## Summary

- They share 32 benchmarks with published results for both. Claude Opus 4 scores higher in 9 categories and Qwen2.5 72B Instruct in 0 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in math, where Claude Opus 4 leads 40.2 to 18.1.
- The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 64.4% for Claude Opus 4 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 $15 / $75 for Claude Opus 4.
- Claude Opus 4 accepts more context: 200K tokens versus 131K.
- Qwen2.5 72B Instruct has downloadable open weights; the other is API-only.

## Snapshot

| | Claude Opus 4 | Qwen2.5 72B Instruct |
|---|---|---|
| Provider | Anthropic | Alibaba (Qwen) |
| Noometry Index | 42.9 | 31.5 |
| Rank | 93 | 259 |
| Context | 200K | 131K |
| Input $/M | $15 | $1.40 |
| Output $/M | $75 | $5.60 |
| Weights | Proprietary | Open |

## Coding

- Claude Opus 4: 47.8 (#55)
- Qwen2.5 72B Instruct: 32.8 (#256)

| Benchmark | Claude Opus 4 | Qwen2.5 72B Instruct |
|---|---|---|
| WeirdML | 43.7% | 16% |
| LMArena Coding | 1442 | 1293 |
| SWE-bench Verified | 70.7% | — |
| SWE-bench Verified (bash only) | 67.6% | — |
| Aider Polyglot | 72% | — |
| GSO | 6.9% | — |
| BigCodeBench Instruct | — | 45.8% |
| BigCodeBench Complete | — | 55.9% |
| AlgoTune | 1.33 | — |

## Agentic & Tool Use

- Claude Opus 4: 34.5 (#39)
- Qwen2.5 72B Instruct: 22.0 (#124)

| Benchmark | Claude Opus 4 | Qwen2.5 72B Instruct |
|---|---|---|
| METR Time Horizons | 63.9% | 35.8% |
| TheAgentCompany | — | 5.7% |
| Cybench | 38% | — |
| DeepResearch Bench | 46.8% | — |
| BALROG | — | 16.2% |
| LMArena Search | 1127 | — |

## Reasoning

- Claude Opus 4: 27.4 (#104)
- Qwen2.5 72B Instruct: 21.4 (#197)

| Benchmark | Claude Opus 4 | Qwen2.5 72B Instruct |
|---|---|---|
| LMArena Hard Prompts | 1399 | 1271 |
| DTBench | 81.6% | 62.9% |
| LMCA | 37.4% | 13.4% |
| Epoch Capabilities Index | 142.67 | 129.01 |
| ForecastBench | 61.1 | 57.5 |
| ARC-AGI-2 | 8.6% | — |
| SimpleBench | 58.8% | — |
| Kagi LLM Benchmark | 74.3% | — |
| ARC-AGI-1 | 35.7% | — |
| CritPt | 0.3% | — |
| EnigmaEval | 5.6% | — |
| BIG-Bench Hard | — | 79.8% |
| HellaSwag | — | 84.8% |
| PIQA | — | 82.6% |
| WinoGrande | — | 82.3% |

## Math

- Claude Opus 4: 40.2 (#80)
- Qwen2.5 72B Instruct: 18.1 (#281)

| Benchmark | Claude Opus 4 | Qwen2.5 72B Instruct |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 64.4% | 8.1% |
| Omni-MATH | 61.6% | 33% |
| LMArena Math | 1391 | 1283 |
| MATH Level 5 | 85% | 63.2% |
| FrontierMath (Feb 2025 set) | 4.5% | — |
| FrontierMath Tier 4 (v1) | 4.2% | — |

## Knowledge

- Claude Opus 4: 43.9 (#82)
- Qwen2.5 72B Instruct: 26.9 (#247)

| Benchmark | Claude Opus 4 | Qwen2.5 72B Instruct |
|---|---|---|
| GPQA Diamond | 76.3% | 49.1% |
| MMLU-Pro | 87.5% | 63.1% |
| Confabulations | 15.9% | 19.1% |
| GPQA (HELM) | 70.8% | 42.6% |
| LMArena Expert | 1384 | 1245 |
| Humanity's Last Exam | 10.7% | — |
| Vectara Hallucination Rate | 12% | — |
| ARC (AI2) Challenge | — | 94.5% |
| MMLU | — | 85.3% |
| TriviaQA | — | 71.9% |

## Multimodal

- Claude Opus 4: 30.9 (#101)
- Qwen2.5 72B Instruct: —

| Benchmark | Claude Opus 4 | Qwen2.5 72B Instruct |
|---|---|---|
| LMArena Vision | 1192 | — |
| GeoBench | 49% | — |
| VPCT | 38% | — |

## Multilingual

- Claude Opus 4: 49.0 (#133)
- Qwen2.5 72B Instruct: 41.2 (#208)

| Benchmark | Claude Opus 4 | Qwen2.5 72B Instruct |
|---|---|---|
| LMArena Non-English | 1362 | 1252 |
| LMArena Chinese | 1386 | 1272 |
| LMArena French | 1370 | 1279 |
| LMArena German | 1391 | 1234 |
| LMArena Japanese | 1332 | 1180 |
| LMArena Korean | 1318 | 1188 |
| LMArena Russian | 1392 | 1264 |
| LMArena Spanish | 1388 | 1255 |

## Instruction Following

- Claude Opus 4: 77.3 (#26)
- Qwen2.5 72B Instruct: 65.7 (#216)

| Benchmark | Claude Opus 4 | Qwen2.5 72B Instruct |
|---|---|---|
| IFEval | 91.8% | 80.6% |
| LMArena Instruction Following | 1406 | 1254 |

## Long Context

- Claude Opus 4: 39.5 (#167)
- Qwen2.5 72B Instruct: 38.9 (#183)

| Benchmark | Claude Opus 4 | Qwen2.5 72B Instruct |
|---|---|---|
| LMArena Longer Query | 1422 | 1282 |
| Fiction.LiveBench | 61.1% | — |

## Writing & Preference

- Claude Opus 4: 61.4 (#85)
- Qwen2.5 72B Instruct: 46.9 (#210)

| Benchmark | Claude Opus 4 | Qwen2.5 72B Instruct |
|---|---|---|
| LMArena Text | 1377 | 1269 |
| LMArena Creative Writing | 1386 | 1221 |
| WildBench | 85.2% | 80.2% |
| LMArena Multi-Turn | 1396 | 1272 |
| Short-Story Creative Writing | 83.6% | — |
| EQ-Bench Creative Writing | 1580 | — |

## FAQ

### Is Claude Opus 4 better than Qwen2.5 72B Instruct?

Claude Opus 4 is the stronger model overall, scoring 42.9 to 31.5 on the Noometry Index. Qwen2.5 72B Instruct costs 12× less per token, which makes it the better buy when Claude Opus 4's lead doesn't matter for your workload.

### Which is cheaper, Claude Opus 4 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 Opus 4 lists at $15 and $75.

### Is Claude Opus 4 or Qwen2.5 72B Instruct better for coding?

Claude Opus 4 scores higher on coding benchmarks: 47.8 versus 32.8 in the Noometry coding category.

### Which has the bigger context window?

Claude Opus 4 does, with 200K tokens against 131K.

### How many benchmarks do Claude Opus 4 and Qwen2.5 72B Instruct share?

32 benchmarks have published results for both models. Claude Opus 4 has 56 scored results on Noometry and Qwen2.5 72B Instruct has 43.