# GPT-5.2 Pro vs Step 5 Preview

> GPT-5.2 Pro is the stronger model overall, scoring 52.3 to 47.9 on the Noometry Index. Step 5 Preview costs 41× less per token, which makes it the better buy when GPT-5.2 Pro's lead doesn't matter for your workload.

- Canonical page: https://noometry.com/compare/gpt-5-2-pro-vs-step-5-preview
- Last updated: 2026-10-11
- Shared benchmarks: 0

## Summary

- The widest gap is in math, where GPT-5.2 Pro leads 65.3 to 46.1.
- Step 5 Preview is cheaper at $1 / $2.70 per million input/output tokens, against $21 / $168 for GPT-5.2 Pro.
- Step 5 Preview accepts more context: 1.02M tokens versus 400K.

## Snapshot

| | GPT-5.2 Pro | Step 5 Preview |
|---|---|---|
| Provider | OpenAI | StepFun |
| Noometry Index | 52.3 | 47.9 |
| Rank | 39 | 58 |
| Context | 400K | 1.02M |
| Input $/M | $21 | $1 |
| Output $/M | $168 | $2.70 |
| Weights | Proprietary | Proprietary |

## Coding

- GPT-5.2 Pro: —
- Step 5 Preview: 51.8 (#36)

| Benchmark | GPT-5.2 Pro | Step 5 Preview |
|---|---|---|
| LMArena WebDev | — | 1564 |
| SciCode | — | 58.9% |
| LMArena Coding | — | 1480 |

## Reasoning

- GPT-5.2 Pro: 51.5 (#33)
- Step 5 Preview: 40.0 (#57)

| Benchmark | GPT-5.2 Pro | Step 5 Preview |
|---|---|---|
| ARC-AGI-2 | 54.2% | — |
| SimpleBench | 57.4% | — |
| NYT Connections (extended) | 79.3% | — |
| ARC-AGI-1 | 90.5% | — |
| CritPt | — | 20.9% |
| LMArena Hard Prompts | — | 1465 |
| Epoch Capabilities Index | 155.4 | — |

## Math

- GPT-5.2 Pro: 65.3 (#29)
- Step 5 Preview: 46.1 (#72)

| Benchmark | GPT-5.2 Pro | Step 5 Preview |
|---|---|---|
| FrontierMath (Tiers 1-3) | 74% | — |
| FrontierMath Tier 4 | 46% | — |
| ProofBench | — | 42% |
| LMArena Math | — | 1470 |
| FrontierMath Tier 4 (v1) | 31.3% | — |

## Knowledge

- GPT-5.2 Pro: —
- Step 5 Preview: 41.2 (#112)

| Benchmark | GPT-5.2 Pro | Step 5 Preview |
|---|---|---|
| LMArena Expert | — | 1470 |

## Multimodal

- GPT-5.2 Pro: —
- Step 5 Preview: 41.0 (#41)

| Benchmark | GPT-5.2 Pro | Step 5 Preview |
|---|---|---|
| LMArena Vision | — | 1267 |

## Multilingual

- GPT-5.2 Pro: —
- Step 5 Preview: 53.8 (#54)

| Benchmark | GPT-5.2 Pro | Step 5 Preview |
|---|---|---|
| LMArena Non-English | — | 1432 |
| LMArena Chinese | — | 1519 |
| LMArena Russian | — | 1436 |
| LMArena Spanish | — | 1439 |

## Instruction Following

- GPT-5.2 Pro: —
- Step 5 Preview: 76.0 (#50)

| Benchmark | GPT-5.2 Pro | Step 5 Preview |
|---|---|---|
| LMArena Instruction Following | — | 1444 |

## Long Context

- GPT-5.2 Pro: —
- Step 5 Preview: 44.5 (#64)

| Benchmark | GPT-5.2 Pro | Step 5 Preview |
|---|---|---|
| LMArena Longer Query | — | 1455 |

## Writing & Preference

- GPT-5.2 Pro: —
- Step 5 Preview: 62.8 (#71)

| Benchmark | GPT-5.2 Pro | Step 5 Preview |
|---|---|---|
| LMArena Text | — | 1442 |
| LMArena Creative Writing | — | 1410 |
| LMArena Multi-Turn | — | 1447 |

## FAQ

### Is GPT-5.2 Pro better than Step 5 Preview?

GPT-5.2 Pro is the stronger model overall, scoring 52.3 to 47.9 on the Noometry Index. Step 5 Preview costs 41× less per token, which makes it the better buy when GPT-5.2 Pro's lead doesn't matter for your workload.

### Which is cheaper, GPT-5.2 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; GPT-5.2 Pro lists at $21 and $168.

### Which has the bigger context window?

Step 5 Preview does, with 1.02M tokens against 400K.

### How many benchmarks do GPT-5.2 Pro and Step 5 Preview share?

0 benchmarks have published results for both models. GPT-5.2 Pro has 8 scored results on Noometry and Step 5 Preview has 18.