# Qwen2-72B vs Step 5 Preview

> Step 5 Preview is the stronger model overall, scoring 47.9 to 30.0 on the Noometry Index.

- Canonical page: https://noometry.com/compare/qwen2-72b-vs-step-5-preview
- Last updated: 2026-10-10
- Shared benchmarks: 13

## Summary

- They share 13 benchmarks with published results for both. Qwen2-72B scores higher in 0 categories and Step 5 Preview in 8 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in coding, where Step 5 Preview leads 51.8 to 29.1.
- Qwen2-72B has downloadable open weights; the other is API-only.

## Snapshot

| | Qwen2-72B | Step 5 Preview |
|---|---|---|
| Provider | Alibaba (Qwen) | StepFun |
| Noometry Index | 30.0 | 47.9 |
| Rank | 300 | 58 |
| Context | — | 1.02M |
| Input $/M | — | $1 |
| Output $/M | — | $2.70 |
| Weights | Open | Proprietary |

## Coding

- Qwen2-72B: 29.1 (#310)
- Step 5 Preview: 51.8 (#36)

| Benchmark | Qwen2-72B | Step 5 Preview |
|---|---|---|
| LMArena Coding | 1196 | 1480 |
| LMArena WebDev | — | 1564 |
| SciCode | — | 58.9% |
| WeirdML | 11.3% | — |
| BigCodeBench Instruct | 38.5% | — |
| BigCodeBench Complete | 54% | — |

## Agentic & Tool Use

- Qwen2-72B: 17.0 (#146)
- Step 5 Preview: —

| Benchmark | Qwen2-72B | Step 5 Preview |
|---|---|---|
| TheAgentCompany | 1.1% | — |
| METR Time Horizons | 29.9% | — |

## Reasoning

- Qwen2-72B: 23.2 (#181)
- Step 5 Preview: 40.0 (#57)

| Benchmark | Qwen2-72B | Step 5 Preview |
|---|---|---|
| LMArena Hard Prompts | 1191 | 1465 |
| CritPt | — | 20.9% |
| Epoch Capabilities Index | 125.28 | — |

## Math

- Qwen2-72B: 30.2 (#236)
- Step 5 Preview: 46.1 (#72)

| Benchmark | Qwen2-72B | Step 5 Preview |
|---|---|---|
| LMArena Math | 1235 | 1470 |
| ProofBench | — | 42% |
| MATH Level 5 | 39.1% | — |

## Knowledge

- Qwen2-72B: 21.2 (#275)
- Step 5 Preview: 41.2 (#112)

| Benchmark | Qwen2-72B | Step 5 Preview |
|---|---|---|
| LMArena Expert | 1171 | 1470 |
| GPQA Diamond | 40.8% | — |
| MMLU | 82.4% | — |

## Multimodal

- Qwen2-72B: —
- Step 5 Preview: 41.0 (#41)

| Benchmark | Qwen2-72B | Step 5 Preview |
|---|---|---|
| LMArena Vision | — | 1267 |

## Multilingual

- Qwen2-72B: 35.9 (#244)
- Step 5 Preview: 53.8 (#54)

| Benchmark | Qwen2-72B | Step 5 Preview |
|---|---|---|
| LMArena Non-English | 1176 | 1432 |
| LMArena Chinese | 1240 | 1519 |
| LMArena Russian | 1169 | 1436 |
| LMArena Spanish | 1169 | 1439 |
| LMArena French | 1170 | — |
| LMArena German | 1151 | — |
| LMArena Japanese | 1111 | — |
| LMArena Korean | 1083 | — |

## Instruction Following

- Qwen2-72B: 61.7 (#241)
- Step 5 Preview: 76.0 (#50)

| Benchmark | Qwen2-72B | Step 5 Preview |
|---|---|---|
| LMArena Instruction Following | 1181 | 1444 |

## Long Context

- Qwen2-72B: 36.1 (#235)
- Step 5 Preview: 44.5 (#64)

| Benchmark | Qwen2-72B | Step 5 Preview |
|---|---|---|
| LMArena Longer Query | 1192 | 1455 |

## Writing & Preference

- Qwen2-72B: 40.8 (#241)
- Step 5 Preview: 62.8 (#71)

| Benchmark | Qwen2-72B | Step 5 Preview |
|---|---|---|
| LMArena Text | 1203 | 1442 |
| LMArena Creative Writing | 1181 | 1410 |
| LMArena Multi-Turn | 1196 | 1447 |

## FAQ

### Is Qwen2-72B better than Step 5 Preview?

Step 5 Preview is the stronger model overall, scoring 47.9 to 30.0 on the Noometry Index.

### Is Qwen2-72B or Step 5 Preview better for coding?

Step 5 Preview scores higher on coding benchmarks: 51.8 versus 29.1 in the Noometry coding category.

### How many benchmarks do Qwen2-72B and Step 5 Preview share?

13 benchmarks have published results for both models. Qwen2-72B has 26 scored results on Noometry and Step 5 Preview has 18.