# Qwen3 32B vs Step 5 Preview

> Step 5 Preview is the stronger model overall, scoring 43.8 to 38.8 on the Noometry Index.

- Canonical page: https://noometry.com/compare/qwen3-32b-vs-step-5-preview
- Last updated: 2026-10-10
- Shared benchmarks: 2

## Summary

- They share 2 benchmarks with published results for both. Qwen3 32B scores higher in 0 categories and Step 5 Preview in 2 categories; 2 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where Step 5 Preview leads 38.0 to 19.6.
- The biggest single-benchmark swing is SciCode: 35.4% for Qwen3 32B and 58.9% for Step 5 Preview.
- Qwen3 32B is cheaper at $0.70 / $2.80 per million input/output tokens, against $1 / $2.70 for Step 5 Preview.
- Step 5 Preview accepts more context: 1.02M tokens versus 131K.
- Qwen3 32B has downloadable open weights; the other is API-only.

## Snapshot

| | Qwen3 32B | Step 5 Preview |
|---|---|---|
| Provider | Alibaba (Qwen) | StepFun |
| Noometry Index | 38.8 | 43.8 |
| Rank | 166 | 78 |
| Context | 131K | 1.02M |
| Input $/M | $0.70 | $1 |
| Output $/M | $2.80 | $2.70 |
| Weights | Open | Proprietary |

## Coding

- Qwen3 32B: 37.3 (#184)
- Step 5 Preview: 50.6 (#40)

| Benchmark | Qwen3 32B | Step 5 Preview |
|---|---|---|
| SciCode | 35.4% | 58.9% |
| Aider Polyglot | 40% | — |
| LMArena WebDev | — | 1569 |
| LMArena Coding | 1358 | — |

## Agentic & Tool Use

- Qwen3 32B: 32.3 (#58)
- Step 5 Preview: —

| Benchmark | Qwen3 32B | Step 5 Preview |
|---|---|---|
| Berkeley Function Calling Leaderboard | 48.7% | — |

## Reasoning

- Qwen3 32B: 19.6 (#233)
- Step 5 Preview: 38.0 (#58)

| Benchmark | Qwen3 32B | Step 5 Preview |
|---|---|---|
| CritPt | 0.3% | 20.9% |
| Kagi LLM Benchmark | 54.9% | — |
| Chess Puzzles | 5% | — |
| LMArena Hard Prompts | 1335 | — |
| DTBench | 67.5% | — |
| LMCA | 17.3% | — |
| Epoch Capabilities Index | 138.51 | — |

## Math

- Qwen3 32B: 37.9 (#89)
- Step 5 Preview: —

| Benchmark | Qwen3 32B | Step 5 Preview |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 66.9% | — |
| LMArena Math | 1399 | — |

## Knowledge

- Qwen3 32B: 40.0 (#112)
- Step 5 Preview: —

| Benchmark | Qwen3 32B | Step 5 Preview |
|---|---|---|
| GPQA Diamond | 65.7% | — |
| Vectara Hallucination Rate | 5.9% | — |
| LMArena Expert | 1362 | — |

## Multilingual

- Qwen3 32B: 45.8 (#162)
- Step 5 Preview: —

| Benchmark | Qwen3 32B | Step 5 Preview |
|---|---|---|
| LMArena Non-English | 1317 | — |
| LMArena Chinese | 1357 | — |
| LMArena German | 1341 | — |
| LMArena Russian | 1311 | — |

## Instruction Following

- Qwen3 32B: 69.1 (#174)
- Step 5 Preview: —

| Benchmark | Qwen3 32B | Step 5 Preview |
|---|---|---|
| LMArena Instruction Following | 1305 | — |

## Long Context

- Qwen3 32B: 43.7 (#83)
- Step 5 Preview: —

| Benchmark | Qwen3 32B | Step 5 Preview |
|---|---|---|
| Fiction.LiveBench | 74.2% | — |
| LMArena Longer Query | 1327 | — |

## Writing & Preference

- Qwen3 32B: 53.1 (#158)
- Step 5 Preview: —

| Benchmark | Qwen3 32B | Step 5 Preview |
|---|---|---|
| LMArena Text | 1340 | — |
| LMArena Creative Writing | 1297 | — |
| LMArena Multi-Turn | 1331 | — |

## FAQ

### Is Qwen3 32B better than Step 5 Preview?

Step 5 Preview is the stronger model overall, scoring 43.8 to 38.8 on the Noometry Index.

### Which is cheaper, Qwen3 32B or Step 5 Preview?

Qwen3 32B is cheaper. It lists at $0.70 per million input tokens and $2.80 per million output tokens; Step 5 Preview lists at $1 and $2.70.

### Is Qwen3 32B or Step 5 Preview better for coding?

Step 5 Preview scores higher on coding benchmarks: 50.6 versus 37.3 in the Noometry coding category.

### Which has the bigger context window?

Step 5 Preview does, with 1.02M tokens against 131K.

### How many benchmarks do Qwen3 32B and Step 5 Preview share?

2 benchmarks have published results for both models. Qwen3 32B has 26 scored results on Noometry and Step 5 Preview has 3.