Model comparison
Grok 3 vs Step 3
Grok 3 and Step 3 score almost the same on the Noometry Index (39.9 vs 40.5), so choose on price, context window or the category you care about most.
Last verified . 16 shared benchmarks.
StepFunSummary
- They share 16 benchmarks with published results for both. Grok 3 scores higher in 6 categories and Step 3 in 2 categories; 7 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where Step 3 leads 28.4 to 13.7.
- Step 3 has downloadable open weights; the other is API-only.
Side by side
| Grok 3 | Step 3 | |
|---|---|---|
| Provider | xAI | StepFun |
| Noometry Index | 39.9 | 40.5 |
| Released | 2025-04-09 | — |
| Weights | Proprietary | Open |
| Context window | — | — |
| Max output | — | — |
| Input $ / M tokens | — | — |
| Output $ / M tokens | — | — |
| Results tracked | 40 | 17 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Grok 3 leads
Grok 3: 41.9 (#115), Step 3: 40.1 (#147)
| Benchmark | Grok 3 | Step 3 |
|---|---|---|
| LMArena Coding | 1432 | 1367 |
| Aider Polyglot | 53.3% | — |
| WeirdML | 37.2% | — |
Agentic & Tool Use Not comparable
Grok 3: 30.5 (#76), Step 3: —
| Benchmark | Grok 3 | Step 3 |
|---|---|---|
| BALROG | 29.5% | — |
Reasoning Step 3 leads
Grok 3: 13.7 (#333), Step 3: 28.4 (#105)
| Benchmark | Grok 3 | Step 3 |
|---|---|---|
| Kagi LLM Benchmark | 61.3% | 62.3% |
| LMArena Hard Prompts | 1434 | 1355 |
| ARC-AGI-2 | 0% | — |
| SimpleBench | 36.1% | — |
| ARC-AGI-1 | 5.5% | — |
| Epoch Capabilities Index | 138.33 | — |
Math Too close to call
Grok 3: 38.0 (#145), Step 3: 37.6 (#148)
| Benchmark | Grok 3 | Step 3 |
|---|---|---|
| LMArena Math | 1391 | 1366 |
| OTIS Mock AIME 2024-2025 | 55.6% | — |
| Omni-MATH | 46.4% | — |
| MATH Level 5 | 88.7% | — |
| FrontierMath (Feb 2025 set) | 3.8% | — |
| FrontierMath Tier 4 (v1) | 0% | — |
Knowledge Grok 3 leads
Grok 3: 46.2 (#82), Step 3: 36.8 (#164)
| Benchmark | Grok 3 | Step 3 |
|---|---|---|
| LMArena Expert | 1421 | 1333 |
| GPQA Diamond | 75.8% | — |
| MMLU-Pro | 78.8% | — |
| Confabulations | 14.2% | — |
| Vectara Hallucination Rate | 5.8% | — |
| GPQA (HELM) | 65% | — |
Multimodal Not comparable
Grok 3: —, Step 3: 35.5 (#86)
| Benchmark | Grok 3 | Step 3 |
|---|---|---|
| LMArena Vision | — | 1177 |
Multilingual Grok 3 leads
Grok 3: 52.3 (#87), Step 3: 46.3 (#159)
| Benchmark | Grok 3 | Step 3 |
|---|---|---|
| LMArena Non-English | 1410 | 1327 |
| LMArena Chinese | 1448 | 1397 |
| LMArena German | 1431 | 1371 |
| LMArena Korean | 1373 | 1269 |
| LMArena Russian | 1416 | 1331 |
| LMArena Spanish | 1417 | 1371 |
| LMArena French | 1460 | — |
| LMArena Japanese | 1387 | — |
Instruction Following Grok 3 leads
Grok 3: 75.0 (#73), Step 3: 70.4 (#164)
| Benchmark | Grok 3 | Step 3 |
|---|---|---|
| LMArena Instruction Following | 1409 | 1332 |
| IFEval | 88.4% | — |
Long Context Step 3 leads
Grok 3: 38.7 (#192), Step 3: 40.3 (#157)
| Benchmark | Grok 3 | Step 3 |
|---|---|---|
| LMArena Longer Query | 1439 | 1326 |
| Fiction.LiveBench | 58.3% | — |
Writing & Preference Grok 3 leads
Grok 3: 55.8 (#141), Step 3: 54.3 (#151)
| Benchmark | Grok 3 | Step 3 |
|---|---|---|
| LMArena Text | 1426 | 1350 |
| LMArena Creative Writing | 1414 | 1321 |
| LMArena Multi-Turn | 1425 | 1341 |
| Short-Story Creative Writing | 76.4% | — |
| EQ-Bench Creative Writing | 1186 | — |
| WildBench | 84.9% | — |
Frequently asked questions
Is Grok 3 better than Step 3?
Grok 3 and Step 3 score almost the same on the Noometry Index (39.9 vs 40.5), so choose on price, context window or the category you care about most.
Is Grok 3 or Step 3 better for coding?
Grok 3 scores higher on coding benchmarks: 41.9 versus 40.1 in the Noometry coding category.
How many benchmarks do Grok 3 and Step 3 share?
16 benchmarks have published results for both models. Grok 3 has 40 scored results on Noometry and Step 3 has 17.