# Amazon Nova Pro vs Step 3

> Step 3 is the stronger model overall, scoring 39.9 to 30.5 on the Noometry Index.

- Canonical page: https://noometry.com/compare/amazon-nova-pro-vs-step-3
- Last updated: 2026-10-10
- Shared benchmarks: 16

## Summary

- They share 16 benchmarks with published results for both. Amazon Nova Pro scores higher in 0 categories and Step 3 in 9 categories; 9 gaps are clear of the uncertainty.
- The widest gap is in multimodal, where Step 3 leads 34.9 to 24.5.
- Step 3 has downloadable open weights; the other is API-only.

## Snapshot

| | Amazon Nova Pro | Step 3 |
|---|---|---|
| Provider | Amazon | StepFun |
| Noometry Index | 30.5 | 39.9 |
| Rank | 278 | 145 |
| Context | 300K | — |
| Input $/M | $0.80 | — |
| Output $/M | $3.20 | — |
| Weights | Proprietary | Open |

## Coding

- Amazon Nova Pro: 34.6 (#224)
- Step 3: 39.5 (#144)

| Benchmark | Amazon Nova Pro | Step 3 |
|---|---|---|
| LMArena Coding | 1270 | 1367 |
| LiveBench Coding | 38.1% | — |

## Agentic & Tool Use

- Amazon Nova Pro: 16.6 (#139)
- Step 3: —

| Benchmark | Amazon Nova Pro | Step 3 |
|---|---|---|
| Berkeley Function Calling Leaderboard | 25% | — |
| TheAgentCompany | 1.7% | — |

## Reasoning

- Amazon Nova Pro: 19.2 (#240)
- Step 3: 27.6 (#102)

| Benchmark | Amazon Nova Pro | Step 3 |
|---|---|---|
| LMArena Hard Prompts | 1246 | 1355 |
| Kagi LLM Benchmark | — | 62.3% |
| LiveBench Reasoning | 32.6% | — |
| LiveBench Data Analysis | 48.3% | — |
| Epoch Capabilities Index | 123.75 | — |
| LiveBench | 43.5% | — |

## Math

- Amazon Nova Pro: 26.4 (#240)
- Step 3: 34.9 (#148)

| Benchmark | Amazon Nova Pro | Step 3 |
|---|---|---|
| LMArena Math | 1252 | 1366 |
| Omni-MATH | 24.2% | — |
| LiveBench Math | 38% | — |

## Knowledge

- Amazon Nova Pro: 27.1 (#246)
- Step 3: 36.2 (#161)

| Benchmark | Amazon Nova Pro | Step 3 |
|---|---|---|
| LMArena Expert | 1211 | 1333 |
| Humanity's Last Exam | 4.4% | — |
| MMLU-Pro | 67.3% | — |
| Confabulations | 30.1% | — |
| Vectara Hallucination Rate | 5.1% | — |
| GPQA (HELM) | 44.6% | — |
| MMLU | 82% | — |

## Multimodal

- Amazon Nova Pro: 24.5 (#121)
- Step 3: 34.9 (#81)

| Benchmark | Amazon Nova Pro | Step 3 |
|---|---|---|
| LMArena Vision | 980 | 1176 |

## Multilingual

- Amazon Nova Pro: 39.9 (#218)
- Step 3: 46.5 (#154)

| Benchmark | Amazon Nova Pro | Step 3 |
|---|---|---|
| LMArena Non-English | 1234 | 1327 |
| LMArena Chinese | 1244 | 1396 |
| LMArena German | 1243 | 1371 |
| LMArena Korean | 1203 | 1271 |
| LMArena Russian | 1240 | 1331 |
| LMArena Spanish | 1182 | 1372 |
| LMArena French | 1270 | — |
| LMArena Japanese | 1200 | — |

## Instruction Following

- Amazon Nova Pro: 65.1 (#221)
- Step 3: 70.6 (#159)

| Benchmark | Amazon Nova Pro | Step 3 |
|---|---|---|
| LMArena Instruction Following | 1235 | 1332 |
| LiveBench Instruction Following | 67.1% | — |
| IFEval | 81.5% | — |

## Long Context

- Amazon Nova Pro: 38.1 (#199)
- Step 3: 40.3 (#152)

| Benchmark | Amazon Nova Pro | Step 3 |
|---|---|---|
| LMArena Longer Query | 1255 | 1326 |

## Writing & Preference

- Amazon Nova Pro: 44.1 (#221)
- Step 3: 54.5 (#146)

| Benchmark | Amazon Nova Pro | Step 3 |
|---|---|---|
| LMArena Text | 1259 | 1351 |
| LMArena Creative Writing | 1212 | 1322 |
| LMArena Multi-Turn | 1245 | 1340 |
| Short-Story Creative Writing | 60.5% | — |
| WildBench | 77.7% | — |
| LiveBench Language | 37% | — |

## FAQ

### Is Amazon Nova Pro better than Step 3?

Step 3 is the stronger model overall, scoring 39.9 to 30.5 on the Noometry Index.

### Is Amazon Nova Pro or Step 3 better for coding?

Step 3 scores higher on coding benchmarks: 39.5 versus 34.6 in the Noometry coding category.

### How many benchmarks do Amazon Nova Pro and Step 3 share?

16 benchmarks have published results for both models. Amazon Nova Pro has 38 scored results on Noometry and Step 3 has 17.