# Qwen2.5 7B Instruct vs Step 3

> Step 3 is the stronger model overall, scoring 39.9 to 28.6 on the Noometry Index.

- Canonical page: https://noometry.com/compare/qwen2-5-7b-instruct-vs-step-3
- Last updated: 2026-10-10
- Shared benchmarks: 0

## Summary

- The widest gap is in math, where Step 3 leads 34.9 to 11.8.

## Snapshot

| | Qwen2.5 7B Instruct | Step 3 |
|---|---|---|
| Provider | Alibaba (Qwen) | StepFun |
| Noometry Index | 28.6 | 39.9 |
| Rank | 314 | 145 |
| Context | 131K | — |
| Input $/M | $0.17 | — |
| Output $/M | $0.70 | — |
| Weights | Open | Open |

## Coding

- Qwen2.5 7B Instruct: 35.9 (#204)
- Step 3: 39.5 (#144)

| Benchmark | Qwen2.5 7B Instruct | Step 3 |
|---|---|---|
| BigCodeBench Instruct | 37.6% | — |
| LMArena Coding | — | 1367 |
| BigCodeBench Complete | 46.1% | — |

## Agentic & Tool Use

- Qwen2.5 7B Instruct: 23.5 (#117)
- Step 3: —

| Benchmark | Qwen2.5 7B Instruct | Step 3 |
|---|---|---|
| BALROG | 7.8% | — |

## Reasoning

- Qwen2.5 7B Instruct: 14.3 (#317)
- Step 3: 27.6 (#102)

| Benchmark | Qwen2.5 7B Instruct | Step 3 |
|---|---|---|
| Kagi LLM Benchmark | — | 62.3% |
| Chess Puzzles | 0% | — |
| LMArena Hard Prompts | — | 1355 |
| DTBench | 47.7% | — |
| LMCA | 6.4% | — |
| Epoch Capabilities Index | 118.44 | — |

## Math

- Qwen2.5 7B Instruct: 11.8 (#300)
- Step 3: 34.9 (#148)

| Benchmark | Qwen2.5 7B Instruct | Step 3 |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 2.5% | — |
| Omni-MATH | 29.4% | — |
| LMArena Math | — | 1366 |

## Knowledge

- Qwen2.5 7B Instruct: 17.0 (#281)
- Step 3: 36.2 (#161)

| Benchmark | Qwen2.5 7B Instruct | Step 3 |
|---|---|---|
| GPQA Diamond | 35.5% | — |
| MMLU-Pro | 53.9% | — |
| GPQA (HELM) | 34.1% | — |
| LMArena Expert | — | 1333 |
| MMLU | 72.9% | — |

## Multimodal

- Qwen2.5 7B Instruct: —
- Step 3: 34.9 (#81)

| Benchmark | Qwen2.5 7B Instruct | Step 3 |
|---|---|---|
| LMArena Vision | — | 1176 |

## Multilingual

- Qwen2.5 7B Instruct: —
- Step 3: 46.5 (#154)

| Benchmark | Qwen2.5 7B Instruct | Step 3 |
|---|---|---|
| LMArena Non-English | — | 1327 |
| LMArena Chinese | — | 1396 |
| LMArena German | — | 1371 |
| LMArena Korean | — | 1271 |
| LMArena Russian | — | 1331 |
| LMArena Spanish | — | 1372 |

## Instruction Following

- Qwen2.5 7B Instruct: 63.4 (#226)
- Step 3: 70.6 (#159)

| Benchmark | Qwen2.5 7B Instruct | Step 3 |
|---|---|---|
| IFEval | 74.1% | — |
| LMArena Instruction Following | — | 1332 |

## Long Context

- Qwen2.5 7B Instruct: —
- Step 3: 40.3 (#152)

| Benchmark | Qwen2.5 7B Instruct | Step 3 |
|---|---|---|
| LMArena Longer Query | — | 1326 |

## Writing & Preference

- Qwen2.5 7B Instruct: 48.8 (#194)
- Step 3: 54.5 (#146)

| Benchmark | Qwen2.5 7B Instruct | Step 3 |
|---|---|---|
| LMArena Text | — | 1351 |
| LMArena Creative Writing | — | 1322 |
| WildBench | 73.1% | — |
| LMArena Multi-Turn | — | 1340 |

## FAQ

### Is Qwen2.5 7B Instruct better than Step 3?

Step 3 is the stronger model overall, scoring 39.9 to 28.6 on the Noometry Index.

### Is Qwen2.5 7B Instruct or Step 3 better for coding?

Step 3 scores higher on coding benchmarks: 39.5 versus 35.9 in the Noometry coding category.

### How many benchmarks do Qwen2.5 7B Instruct and Step 3 share?

0 benchmarks have published results for both models. Qwen2.5 7B Instruct has 15 scored results on Noometry and Step 3 has 17.