# Claude 3 Opus vs Step 5 Preview

> Step 5 Preview is the stronger model overall, scoring 47.9 to 29.5 on the Noometry Index.

- Canonical page: https://noometry.com/compare/claude-3-opus-vs-step-5-preview
- Last updated: 2026-10-10
- Shared benchmarks: 14

## Summary

- They share 14 benchmarks with published results for both. Claude 3 Opus scores higher in 0 categories and Step 5 Preview in 9 categories; 9 gaps are clear of the uncertainty.
- The widest gap is in math, where Step 5 Preview leads 46.1 to 14.8.

## Snapshot

| | Claude 3 Opus | Step 5 Preview |
|---|---|---|
| Provider | Anthropic | StepFun |
| Noometry Index | 29.5 | 47.9 |
| Rank | 310 | 58 |
| Context | — | 1.02M |
| Input $/M | — | $1 |
| Output $/M | — | $2.70 |
| Weights | Proprietary | Proprietary |

## Coding

- Claude 3 Opus: 32.9 (#267)
- Step 5 Preview: 51.8 (#36)

| Benchmark | Claude 3 Opus | Step 5 Preview |
|---|---|---|
| LMArena Coding | 1264 | 1480 |
| LMArena WebDev | — | 1564 |
| SciCode | — | 58.9% |
| WeirdML | 19.2% | — |
| BigCodeBench Instruct | 45.5% | — |
| LiveBench Coding | 38.6% | — |
| BigCodeBench Complete | 57.4% | — |
| HumanEval+ | 77.4% | — |
| MBPP+ | 73.3% | — |

## Agentic & Tool Use

- Claude 3 Opus: 24.6 (#116)
- Step 5 Preview: —

| Benchmark | Claude 3 Opus | Step 5 Preview |
|---|---|---|
| Cybench | 10% | — |
| METR Time Horizons | 29.5% | — |

## Reasoning

- Claude 3 Opus: 14.6 (#324)
- Step 5 Preview: 40.0 (#57)

| Benchmark | Claude 3 Opus | Step 5 Preview |
|---|---|---|
| LMArena Hard Prompts | 1245 | 1465 |
| SimpleBench | 23.5% | — |
| CritPt | — | 20.9% |
| Chess Puzzles | 5% | — |
| EnigmaEval | 0.8% | — |
| LiveBench Reasoning | 40.6% | — |
| DTBench | 61.6% | — |
| LiveBench Data Analysis | 57.9% | — |
| LMCA | 17% | — |
| Epoch Capabilities Index | 126.91 | — |
| ForecastBench | 58.4 | — |
| LiveBench | 49.2% | — |
| WinoGrande | 88.5% | — |

## Math

- Claude 3 Opus: 14.8 (#299)
- Step 5 Preview: 46.1 (#72)

| Benchmark | Claude 3 Opus | Step 5 Preview |
|---|---|---|
| LMArena Math | 1273 | 1470 |
| OTIS Mock AIME 2024-2025 | 4.7% | — |
| ProofBench | — | 42% |
| LiveBench Math | 43.6% | — |
| MATH Level 5 | 37.5% | — |

## Knowledge

- Claude 3 Opus: 24.5 (#267)
- Step 5 Preview: 41.2 (#112)

| Benchmark | Claude 3 Opus | Step 5 Preview |
|---|---|---|
| LMArena Expert | 1223 | 1470 |
| GPQA Diamond | 47.2% | — |
| SimpleQA Verified | 12.6% | — |
| Confabulations | 22.7% | — |
| MMLU | 84.6% | — |

## Multimodal

- Claude 3 Opus: 27.1 (#116)
- Step 5 Preview: 41.0 (#41)

| Benchmark | Claude 3 Opus | Step 5 Preview |
|---|---|---|
| LMArena Vision | 1023 | 1267 |

## Multilingual

- Claude 3 Opus: 41.4 (#207)
- Step 5 Preview: 53.8 (#54)

| Benchmark | Claude 3 Opus | Step 5 Preview |
|---|---|---|
| LMArena Non-English | 1258 | 1432 |
| LMArena Chinese | 1248 | 1519 |
| LMArena Russian | 1280 | 1436 |
| LMArena Spanish | 1246 | 1439 |
| LMArena French | 1275 | — |
| LMArena German | 1258 | — |
| LMArena Japanese | 1204 | — |
| LMArena Korean | 1187 | — |

## Instruction Following

- Claude 3 Opus: 64.1 (#228)
- Step 5 Preview: 76.0 (#50)

| Benchmark | Claude 3 Opus | Step 5 Preview |
|---|---|---|
| LMArena Instruction Following | 1248 | 1444 |
| LiveBench Instruction Following | 63.9% | — |

## Long Context

- Claude 3 Opus: 38.2 (#202)
- Step 5 Preview: 44.5 (#64)

| Benchmark | Claude 3 Opus | Step 5 Preview |
|---|---|---|
| LMArena Longer Query | 1259 | 1455 |

## Writing & Preference

- Claude 3 Opus: 47.2 (#213)
- Step 5 Preview: 62.8 (#71)

| Benchmark | Claude 3 Opus | Step 5 Preview |
|---|---|---|
| LMArena Text | 1262 | 1442 |
| LMArena Creative Writing | 1235 | 1410 |
| LMArena Multi-Turn | 1275 | 1447 |
| LiveBench Language | 50.4% | — |

## FAQ

### Is Claude 3 Opus better than Step 5 Preview?

Step 5 Preview is the stronger model overall, scoring 47.9 to 29.5 on the Noometry Index.

### Is Claude 3 Opus or Step 5 Preview better for coding?

Step 5 Preview scores higher on coding benchmarks: 51.8 versus 32.9 in the Noometry coding category.

### How many benchmarks do Claude 3 Opus and Step 5 Preview share?

14 benchmarks have published results for both models. Claude 3 Opus has 46 scored results on Noometry and Step 5 Preview has 18.
