# o3-pro vs Step 5 Preview

> Step 5 Preview is the stronger model overall, scoring 47.9 to 42.9 on the Noometry Index.

- Canonical page: https://noometry.com/compare/o3-pro-vs-step-5-preview
- Last updated: 2026-10-11
- Shared benchmarks: 0

## Summary

- The widest gap is in long context, where o3-pro leads 72.2 to 44.5.
- Step 5 Preview is cheaper at $1 / $2.70 per million input/output tokens, against $20 / $80 for o3-pro.
- Step 5 Preview accepts more context: 1.02M tokens versus 200K.

## Snapshot

| | o3-pro | Step 5 Preview |
|---|---|---|
| Provider | OpenAI | StepFun |
| Noometry Index | 42.9 | 47.9 |
| Rank | 105 | 58 |
| Context | 200K | 1.02M |
| Input $/M | $20 | $1 |
| Output $/M | $80 | $2.70 |
| Weights | Proprietary | Proprietary |

## Coding

- o3-pro: 55.5 (#24)
- Step 5 Preview: 51.8 (#36)

| Benchmark | o3-pro | Step 5 Preview |
|---|---|---|
| Aider Polyglot | 84.9% | — |
| LMArena WebDev | — | 1564 |
| SciCode | — | 58.9% |
| WeirdML | 58.2% | — |
| LMArena Coding | — | 1480 |

## Reasoning

- o3-pro: 23.8 (#171)
- Step 5 Preview: 40.0 (#57)

| Benchmark | o3-pro | Step 5 Preview |
|---|---|---|
| ARC-AGI-2 | 4.9% | — |
| Kagi LLM Benchmark | 72.1% | — |
| ARC-AGI-1 | 59.3% | — |
| CritPt | — | 20.9% |
| LMArena Hard Prompts | — | 1465 |
| DTBench | 86.9% | — |
| LMCA | 38.5% | — |
| Epoch Capabilities Index | 147.42 | — |

## Math

- o3-pro: —
- Step 5 Preview: 46.1 (#72)

| Benchmark | o3-pro | Step 5 Preview |
|---|---|---|
| ProofBench | — | 42% |
| LMArena Math | — | 1470 |

## Knowledge

- o3-pro: 29.5 (#238)
- Step 5 Preview: 41.2 (#112)

| Benchmark | o3-pro | Step 5 Preview |
|---|---|---|
| Confabulations | 14.2% | — |
| Vectara Hallucination Rate | 23.3% | — |
| LMArena Expert | — | 1470 |

## Multimodal

- o3-pro: —
- Step 5 Preview: 41.0 (#41)

| Benchmark | o3-pro | Step 5 Preview |
|---|---|---|
| LMArena Vision | — | 1267 |

## Multilingual

- o3-pro: —
- Step 5 Preview: 53.8 (#54)

| Benchmark | o3-pro | Step 5 Preview |
|---|---|---|
| LMArena Non-English | — | 1432 |
| LMArena Chinese | — | 1519 |
| LMArena Russian | — | 1436 |
| LMArena Spanish | — | 1439 |

## Instruction Following

- o3-pro: —
- Step 5 Preview: 76.0 (#50)

| Benchmark | o3-pro | Step 5 Preview |
|---|---|---|
| LMArena Instruction Following | — | 1444 |

## Long Context

- o3-pro: 72.2 (#1)
- Step 5 Preview: 44.5 (#64)

| Benchmark | o3-pro | Step 5 Preview |
|---|---|---|
| Fiction.LiveBench | 97.2% | — |
| LMArena Longer Query | — | 1455 |

## Writing & Preference

- o3-pro: 57.1 (#133)
- Step 5 Preview: 62.8 (#71)

| Benchmark | o3-pro | Step 5 Preview |
|---|---|---|
| LMArena Text | — | 1442 |
| LMArena Creative Writing | — | 1410 |
| Short-Story Creative Writing | 84.4% | — |
| LMArena Multi-Turn | — | 1447 |

## FAQ

### Is o3-pro better than Step 5 Preview?

Step 5 Preview is the stronger model overall, scoring 47.9 to 42.9 on the Noometry Index.

### Which is cheaper, o3-pro or Step 5 Preview?

Step 5 Preview is cheaper. It lists at $1 per million input tokens and $2.70 per million output tokens; o3-pro lists at $20 and $80.

### Is o3-pro or Step 5 Preview better for coding?

o3-pro scores higher on coding benchmarks: 55.5 versus 51.8 in the Noometry coding category.

### Which has the bigger context window?

Step 5 Preview does, with 1.02M tokens against 200K.

### How many benchmarks do o3-pro and Step 5 Preview share?

0 benchmarks have published results for both models. o3-pro has 12 scored results on Noometry and Step 5 Preview has 18.