Model comparison
Olmo 3.1 32b Think vs Step 5 Preview
Step 5 Preview is the stronger model overall, scoring 47.9 to 37.9 on the Noometry Index.
Last verified . 13 shared benchmarks.
Allen Institute for AI (Ai2)
StepFunSummary
- They share 13 benchmarks with published results for both. Olmo 3.1 32b Think scores higher in 0 categories and Step 5 Preview in 8 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in writing & preference, where Step 5 Preview leads 62.8 to 46.2.
- Olmo 3.1 32b Think has downloadable open weights; the other is API-only.
Side by side
| Olmo 3.1 32b Think | Step 5 Preview | |
|---|---|---|
| Provider | Allen Institute for AI (Ai2) | StepFun |
| Noometry Index | 37.9 | 47.9 |
| Released | — | 2026-09-16 |
| Weights | Open | Proprietary |
| Context window | — | 1.02M |
| Max output | — | 66K |
| Input $ / M tokens | — | $1 |
| Output $ / M tokens | — | $2.70 |
| Results tracked | 15 | 18 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Step 5 Preview leads
Olmo 3.1 32b Think: 37.7 (#189), Step 5 Preview: 51.8 (#36)
| Benchmark | Olmo 3.1 32b Think | Step 5 Preview |
|---|---|---|
| LMArena Coding | 1291 | 1480 |
| LMArena WebDev | — | 1564 |
| SciCode | — | 58.9% |
Reasoning Step 5 Preview leads
Olmo 3.1 32b Think: 25.2 (#150), Step 5 Preview: 40.0 (#57)
| Benchmark | Olmo 3.1 32b Think | Step 5 Preview |
|---|---|---|
| LMArena Hard Prompts | 1272 | 1465 |
| CritPt | — | 20.9% |
Math Step 5 Preview leads
Olmo 3.1 32b Think: 36.3 (#168), Step 5 Preview: 46.1 (#72)
| Benchmark | Olmo 3.1 32b Think | Step 5 Preview |
|---|---|---|
| LMArena Math | 1305 | 1470 |
| ProofBench | — | 42% |
Knowledge Step 5 Preview leads
Olmo 3.1 32b Think: 35.7 (#181), Step 5 Preview: 41.2 (#112)
| Benchmark | Olmo 3.1 32b Think | Step 5 Preview |
|---|---|---|
| LMArena Expert | 1295 | 1470 |
Multimodal Not comparable
Olmo 3.1 32b Think: —, Step 5 Preview: 41.0 (#41)
| Benchmark | Olmo 3.1 32b Think | Step 5 Preview |
|---|---|---|
| LMArena Vision | — | 1267 |
Multilingual Step 5 Preview leads
Olmo 3.1 32b Think: 38.1 (#231), Step 5 Preview: 53.8 (#54)
| Benchmark | Olmo 3.1 32b Think | Step 5 Preview |
|---|---|---|
| LMArena Non-English | 1209 | 1432 |
| LMArena Chinese | 1242 | 1519 |
| LMArena Russian | 1193 | 1436 |
| LMArena Spanish | 1289 | 1439 |
| LMArena French | 1260 | — |
| LMArena German | 1262 | — |
Instruction Following Step 5 Preview leads
Olmo 3.1 32b Think: 65.6 (#218), Step 5 Preview: 76.0 (#50)
| Benchmark | Olmo 3.1 32b Think | Step 5 Preview |
|---|---|---|
| LMArena Instruction Following | 1247 | 1444 |
Long Context Step 5 Preview leads
Olmo 3.1 32b Think: 38.6 (#195), Step 5 Preview: 44.5 (#64)
| Benchmark | Olmo 3.1 32b Think | Step 5 Preview |
|---|---|---|
| LMArena Longer Query | 1272 | 1455 |
Writing & Preference Step 5 Preview leads
Olmo 3.1 32b Think: 46.2 (#220), Step 5 Preview: 62.8 (#71)
| Benchmark | Olmo 3.1 32b Think | Step 5 Preview |
|---|---|---|
| LMArena Text | 1272 | 1442 |
| LMArena Creative Writing | 1226 | 1410 |
| LMArena Multi-Turn | 1252 | 1447 |
Frequently asked questions
Is Olmo 3.1 32b Think better than Step 5 Preview?
Step 5 Preview is the stronger model overall, scoring 47.9 to 37.9 on the Noometry Index.
Is Olmo 3.1 32b Think or Step 5 Preview better for coding?
Step 5 Preview scores higher on coding benchmarks: 51.8 versus 37.7 in the Noometry coding category.
How many benchmarks do Olmo 3.1 32b Think and Step 5 Preview share?
13 benchmarks have published results for both models. Olmo 3.1 32b Think has 15 scored results on Noometry and Step 5 Preview has 18.