Model comparison
Nova 2 Lite vs Step 3
Nova 2 Lite and Step 3 score almost the same on the Noometry Index (39.7 vs 40.5), so choose on price, context window or the category you care about most.
Last verified . 15 shared benchmarks.
StepFunSummary
- They share 15 benchmarks with published results for both. Nova 2 Lite scores higher in 5 categories and Step 3 in 3 categories; one gap is clear of the uncertainty.
- The widest gap is in knowledge, where Nova 2 Lite leads 43.0 to 36.8.
- Step 3 has downloadable open weights; the other is API-only.
Side by side
| Nova 2 Lite | Step 3 | |
|---|---|---|
| Provider | Amazon | StepFun |
| Noometry Index | 39.7 | 40.5 |
| Released | 2025-12-01 | — |
| Weights | Proprietary | Open |
| Context window | 1M | — |
| Max output | 64K | — |
| Input $ / M tokens | $0.30 | — |
| Output $ / M tokens | $2.50 | — |
| Results tracked | 19 | 17 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Too close to call
Nova 2 Lite: 40.7 (#134), Step 3: 40.1 (#147)
| Benchmark | Nova 2 Lite | Step 3 |
|---|---|---|
| LMArena Coding | 1385 | 1367 |
Agentic & Tool Use Not comparable
Nova 2 Lite: 24.1 (#121), Step 3: —
| Benchmark | Nova 2 Lite | Step 3 |
|---|---|---|
| Berkeley Function Calling Leaderboard | 27.1% | — |
Reasoning Too close to call
Nova 2 Lite: 27.5 (#118), Step 3: 28.4 (#105)
| Benchmark | Nova 2 Lite | Step 3 |
|---|---|---|
| LMArena Hard Prompts | 1364 | 1355 |
| Kagi LLM Benchmark | — | 62.3% |
Math Too close to call
Nova 2 Lite: 37.5 (#156), Step 3: 37.6 (#148)
| Benchmark | Nova 2 Lite | Step 3 |
|---|---|---|
| LMArena Math | 1359 | 1366 |
Knowledge Nova 2 Lite leads
Nova 2 Lite: 43.0 (#94), Step 3: 36.8 (#164)
| Benchmark | Nova 2 Lite | Step 3 |
|---|---|---|
| LMArena Expert | 1358 | 1333 |
| Vectara Hallucination Rate | 5.1% | — |
Multimodal Not comparable
Nova 2 Lite: —, Step 3: 35.5 (#86)
| Benchmark | Nova 2 Lite | Step 3 |
|---|---|---|
| LMArena Vision | — | 1177 |
Multilingual Too close to call
Nova 2 Lite: 47.1 (#153), Step 3: 46.3 (#159)
| Benchmark | Nova 2 Lite | Step 3 |
|---|---|---|
| LMArena Non-English | 1337 | 1327 |
| LMArena Chinese | 1364 | 1397 |
| LMArena German | 1343 | 1371 |
| LMArena Korean | 1284 | 1269 |
| LMArena Russian | 1343 | 1331 |
| LMArena Spanish | 1373 | 1371 |
| LMArena French | 1381 | — |
| LMArena Japanese | 1271 | — |
Instruction Following Too close to call
Nova 2 Lite: 70.5 (#161), Step 3: 70.4 (#164)
| Benchmark | Nova 2 Lite | Step 3 |
|---|---|---|
| LMArena Instruction Following | 1335 | 1332 |
Long Context Too close to call
Nova 2 Lite: 40.6 (#150), Step 3: 40.3 (#157)
| Benchmark | Nova 2 Lite | Step 3 |
|---|---|---|
| LMArena Longer Query | 1335 | 1326 |
Writing & Preference Too close to call
Nova 2 Lite: 53.9 (#154), Step 3: 54.3 (#151)
| Benchmark | Nova 2 Lite | Step 3 |
|---|---|---|
| LMArena Text | 1362 | 1350 |
| LMArena Creative Writing | 1291 | 1321 |
| LMArena Multi-Turn | 1338 | 1341 |
Frequently asked questions
Is Nova 2 Lite better than Step 3?
Nova 2 Lite and Step 3 score almost the same on the Noometry Index (39.7 vs 40.5), so choose on price, context window or the category you care about most.
Is Nova 2 Lite or Step 3 better for coding?
They score almost the same on coding (40.7 vs 40.1); test both on your own repository before choosing.
How many benchmarks do Nova 2 Lite and Step 3 share?
15 benchmarks have published results for both models. Nova 2 Lite has 19 scored results on Noometry and Step 3 has 17.