Model comparison
Nemotron 3 Ultra vs Step 3
Nemotron 3 Ultra is the stronger model overall, scoring 42.5 to 40.5 on the Noometry Index.
Last verified . 15 shared benchmarks.
NVIDIA
StepFunSummary
- They share 15 benchmarks with published results for both. Nemotron 3 Ultra scores higher in 6 categories and Step 3 in 2 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in knowledge, where Nemotron 3 Ultra leads 52.5 to 36.8.
Side by side
| Nemotron 3 Ultra | Step 3 | |
|---|---|---|
| Provider | NVIDIA | StepFun |
| Noometry Index | 42.5 | 40.5 |
| Released | 2026-06-04 | — |
| Weights | Open | Open |
| Context window | 262K | — |
| Max output | 128K | — |
| Input $ / M tokens | $0.50 | — |
| Output $ / M tokens | $2.20 | — |
| Results tracked | 30 | 17 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Step 3 leads
Nemotron 3 Ultra: 38.1 (#182), Step 3: 40.1 (#147)
| Benchmark | Nemotron 3 Ultra | Step 3 |
|---|---|---|
| LMArena Coding | 1468 | 1367 |
| FrontierCode | 13.6% | — |
| SciCode | 40.3% | — |
| WeirdML | 43.5% | — |
Agentic & Tool Use Not comparable
Nemotron 3 Ultra: 23.4 (#126), Step 3: —
| Benchmark | Nemotron 3 Ultra | Step 3 |
|---|---|---|
| APEX-Agents | 22.7% | — |
Reasoning Nemotron 3 Ultra leads
Nemotron 3 Ultra: 31.0 (#82), Step 3: 28.4 (#105)
| Benchmark | Nemotron 3 Ultra | Step 3 |
|---|---|---|
| LMArena Hard Prompts | 1452 | 1355 |
| Kagi LLM Benchmark | — | 62.3% |
| CritPt | 3.1% | — |
| Chess Puzzles | 12% | — |
| Mystery Game Puzzles | 20% | — |
| DTBench | 90.1% | — |
| LMCA | 36.9% | — |
| Epoch Capabilities Index | 146.17 | — |
Math Step 3 leads
Nemotron 3 Ultra: 35.3 (#187), Step 3: 37.6 (#148)
| Benchmark | Nemotron 3 Ultra | Step 3 |
|---|---|---|
| LMArena Math | 1457 | 1366 |
| OTIS Mock AIME 2024-2025 | 86.7% | — |
| ProofBench | 2% | — |
Knowledge Nemotron 3 Ultra leads
Nemotron 3 Ultra: 52.5 (#61), Step 3: 36.8 (#164)
| Benchmark | Nemotron 3 Ultra | Step 3 |
|---|---|---|
| LMArena Expert | 1472 | 1333 |
| GPQA Diamond | 85.4% | — |
Multimodal Not comparable
Nemotron 3 Ultra: —, Step 3: 35.5 (#86)
| Benchmark | Nemotron 3 Ultra | Step 3 |
|---|---|---|
| LMArena Vision | — | 1177 |
Multilingual Nemotron 3 Ultra leads
Nemotron 3 Ultra: 53.5 (#64), Step 3: 46.3 (#159)
| Benchmark | Nemotron 3 Ultra | Step 3 |
|---|---|---|
| LMArena Non-English | 1427 | 1327 |
| LMArena Chinese | 1497 | 1397 |
| LMArena German | 1471 | 1371 |
| LMArena Korean | 1386 | 1269 |
| LMArena Russian | 1417 | 1331 |
| LMArena Spanish | 1454 | 1371 |
| LMArena French | 1472 | — |
Instruction Following Nemotron 3 Ultra leads
Nemotron 3 Ultra: 74.7 (#88), Step 3: 70.4 (#164)
| Benchmark | Nemotron 3 Ultra | Step 3 |
|---|---|---|
| LMArena Instruction Following | 1418 | 1332 |
Long Context Nemotron 3 Ultra leads
Nemotron 3 Ultra: 43.9 (#81), Step 3: 40.3 (#157)
| Benchmark | Nemotron 3 Ultra | Step 3 |
|---|---|---|
| LMArena Longer Query | 1437 | 1326 |
Writing & Preference Nemotron 3 Ultra leads
Nemotron 3 Ultra: 66.3 (#36), Step 3: 54.3 (#151)
| Benchmark | Nemotron 3 Ultra | Step 3 |
|---|---|---|
| LMArena Text | 1445 | 1350 |
| LMArena Creative Writing | 1398 | 1321 |
| LMArena Multi-Turn | 1411 | 1341 |
| EQ-Bench Creative Writing | 1692 | — |
Frequently asked questions
Is Nemotron 3 Ultra better than Step 3?
Nemotron 3 Ultra is the stronger model overall, scoring 42.5 to 40.5 on the Noometry Index.
Is Nemotron 3 Ultra or Step 3 better for coding?
Step 3 scores higher on coding benchmarks: 40.1 versus 38.1 in the Noometry coding category.
How many benchmarks do Nemotron 3 Ultra and Step 3 share?
15 benchmarks have published results for both models. Nemotron 3 Ultra has 30 scored results on Noometry and Step 3 has 17.