# Olmo 3.1 32b Think vs Step 3

> Step 3 is the stronger model overall, scoring 39.9 to 37.3 on the Noometry Index.

- Canonical page: https://noometry.com/compare/olmo-3-1-32b-think-vs-step-3
- Last updated: 2026-10-10
- Shared benchmarks: 14

## Summary

- They share 14 benchmarks with published results for both. Olmo 3.1 32b Think scores higher in 0 categories and Step 3 in 8 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in multilingual, where Step 3 leads 46.5 to 38.3.

## Snapshot

| | Olmo 3.1 32b Think | Step 3 |
|---|---|---|
| Provider | Allen Institute for AI (Ai2) | StepFun |
| Noometry Index | 37.3 | 39.9 |
| Rank | 191 | 145 |
| Context | — | — |
| Input $/M | — | — |
| Output $/M | — | — |
| Weights | Open | Open |

## Coding

- Olmo 3.1 32b Think: 37.1 (#186)
- Step 3: 39.5 (#144)

| Benchmark | Olmo 3.1 32b Think | Step 3 |
|---|---|---|
| LMArena Coding | 1292 | 1367 |

## Reasoning

- Olmo 3.1 32b Think: 24.1 (#151)
- Step 3: 27.6 (#102)

| Benchmark | Olmo 3.1 32b Think | Step 3 |
|---|---|---|
| LMArena Hard Prompts | 1272 | 1355 |
| Kagi LLM Benchmark | — | 62.3% |

## Math

- Olmo 3.1 32b Think: 33.6 (#170)
- Step 3: 34.9 (#148)

| Benchmark | Olmo 3.1 32b Think | Step 3 |
|---|---|---|
| LMArena Math | 1305 | 1366 |

## Knowledge

- Olmo 3.1 32b Think: 35.0 (#178)
- Step 3: 36.2 (#161)

| Benchmark | Olmo 3.1 32b Think | Step 3 |
|---|---|---|
| LMArena Expert | 1294 | 1333 |

## Multimodal

- Olmo 3.1 32b Think: —
- Step 3: 34.9 (#81)

| Benchmark | Olmo 3.1 32b Think | Step 3 |
|---|---|---|
| LMArena Vision | — | 1176 |

## Multilingual

- Olmo 3.1 32b Think: 38.3 (#226)
- Step 3: 46.5 (#154)

| Benchmark | Olmo 3.1 32b Think | Step 3 |
|---|---|---|
| LMArena Non-English | 1210 | 1327 |
| LMArena Chinese | 1241 | 1396 |
| LMArena German | 1260 | 1371 |
| LMArena Russian | 1193 | 1331 |
| LMArena Spanish | 1288 | 1372 |
| LMArena French | 1262 | — |
| LMArena Korean | — | 1271 |

## Instruction Following

- Olmo 3.1 32b Think: 65.8 (#213)
- Step 3: 70.6 (#159)

| Benchmark | Olmo 3.1 32b Think | Step 3 |
|---|---|---|
| LMArena Instruction Following | 1246 | 1332 |

## Long Context

- Olmo 3.1 32b Think: 38.6 (#190)
- Step 3: 40.3 (#152)

| Benchmark | Olmo 3.1 32b Think | Step 3 |
|---|---|---|
| LMArena Longer Query | 1272 | 1326 |

## Writing & Preference

- Olmo 3.1 32b Think: 46.4 (#215)
- Step 3: 54.5 (#146)

| Benchmark | Olmo 3.1 32b Think | Step 3 |
|---|---|---|
| LMArena Text | 1271 | 1351 |
| LMArena Creative Writing | 1226 | 1322 |
| LMArena Multi-Turn | 1252 | 1340 |

## FAQ

### Is Olmo 3.1 32b Think better than Step 3?

Step 3 is the stronger model overall, scoring 39.9 to 37.3 on the Noometry Index.

### Is Olmo 3.1 32b Think or Step 3 better for coding?

Step 3 scores higher on coding benchmarks: 39.5 versus 37.1 in the Noometry coding category.

### How many benchmarks do Olmo 3.1 32b Think and Step 3 share?

14 benchmarks have published results for both models. Olmo 3.1 32b Think has 15 scored results on Noometry and Step 3 has 17.