# Amazon Nova Experimental Chat 10 20 vs Step 3

> Amazon Nova Experimental Chat 10 20 is the stronger model overall, scoring 42.1 to 40.5 on the Noometry Index.

- Canonical page: https://noometry.com/compare/amazon-nova-experimental-chat-10-20-vs-step-3
- Last updated: 2026-10-10
- Shared benchmarks: 15

## Summary

- They share 15 benchmarks with published results for both. Amazon Nova Experimental Chat 10 20 scores higher in 7 categories and Step 3 in 1 category; 7 gaps are clear of the uncertainty.
- The widest gap is in multilingual, where Amazon Nova Experimental Chat 10 20 leads 49.5 to 46.3.
- Step 3 has downloadable open weights; the other is API-only.

## Snapshot

| | Amazon Nova Experimental Chat 10 20 | Step 3 |
|---|---|---|
| Provider | Amazon | StepFun |
| Noometry Index | 42.1 | 40.5 |
| Rank | 118 | 149 |
| Context | — | — |
| Input $/M | — | — |
| Output $/M | — | — |
| Weights | Proprietary | Open |

## Coding

- Amazon Nova Experimental Chat 10 20: 41.6 (#123)
- Step 3: 40.1 (#147)

| Benchmark | Amazon Nova Experimental Chat 10 20 | Step 3 |
|---|---|---|
| LMArena Coding | 1412 | 1367 |

## Reasoning

- Amazon Nova Experimental Chat 10 20: 28.4 (#106)
- Step 3: 28.4 (#105)

| Benchmark | Amazon Nova Experimental Chat 10 20 | Step 3 |
|---|---|---|
| LMArena Hard Prompts | 1396 | 1355 |
| Kagi LLM Benchmark | — | 62.3% |

## Math

- Amazon Nova Experimental Chat 10 20: 39.0 (#119)
- Step 3: 37.6 (#148)

| Benchmark | Amazon Nova Experimental Chat 10 20 | Step 3 |
|---|---|---|
| LMArena Math | 1425 | 1366 |

## Knowledge

- Amazon Nova Experimental Chat 10 20: 38.5 (#144)
- Step 3: 36.8 (#164)

| Benchmark | Amazon Nova Experimental Chat 10 20 | Step 3 |
|---|---|---|
| LMArena Expert | 1386 | 1333 |

## Multimodal

- Amazon Nova Experimental Chat 10 20: —
- Step 3: 35.5 (#86)

| Benchmark | Amazon Nova Experimental Chat 10 20 | Step 3 |
|---|---|---|
| LMArena Vision | — | 1177 |

## Multilingual

- Amazon Nova Experimental Chat 10 20: 49.5 (#131)
- Step 3: 46.3 (#159)

| Benchmark | Amazon Nova Experimental Chat 10 20 | Step 3 |
|---|---|---|
| LMArena Non-English | 1372 | 1327 |
| LMArena Chinese | 1404 | 1397 |
| LMArena German | 1391 | 1371 |
| LMArena Korean | 1324 | 1269 |
| LMArena Russian | 1371 | 1331 |
| LMArena Spanish | 1382 | 1371 |
| LMArena French | 1425 | — |
| LMArena Japanese | 1360 | — |

## Instruction Following

- Amazon Nova Experimental Chat 10 20: 72.1 (#141)
- Step 3: 70.4 (#164)

| Benchmark | Amazon Nova Experimental Chat 10 20 | Step 3 |
|---|---|---|
| LMArena Instruction Following | 1365 | 1332 |

## Long Context

- Amazon Nova Experimental Chat 10 20: 41.7 (#133)
- Step 3: 40.3 (#157)

| Benchmark | Amazon Nova Experimental Chat 10 20 | Step 3 |
|---|---|---|
| LMArena Longer Query | 1370 | 1326 |

## Writing & Preference

- Amazon Nova Experimental Chat 10 20: 56.6 (#136)
- Step 3: 54.3 (#151)

| Benchmark | Amazon Nova Experimental Chat 10 20 | Step 3 |
|---|---|---|
| LMArena Text | 1394 | 1350 |
| LMArena Creative Writing | 1318 | 1321 |
| LMArena Multi-Turn | 1364 | 1341 |

## FAQ

### Is Amazon Nova Experimental Chat 10 20 better than Step 3?

Amazon Nova Experimental Chat 10 20 is the stronger model overall, scoring 42.1 to 40.5 on the Noometry Index.

### Is Amazon Nova Experimental Chat 10 20 or Step 3 better for coding?

Amazon Nova Experimental Chat 10 20 scores higher on coding benchmarks: 41.6 versus 40.1 in the Noometry coding category.

### How many benchmarks do Amazon Nova Experimental Chat 10 20 and Step 3 share?

15 benchmarks have published results for both models. Amazon Nova Experimental Chat 10 20 has 17 scored results on Noometry and Step 3 has 17.