# Grok-3 mini vs Step 3

> Grok-3 mini is the stronger model overall, scoring 41.0 to 39.9 on the Noometry Index.

- Canonical page: https://noometry.com/compare/grok-3-mini-vs-step-3
- Last updated: 2026-10-10
- Shared benchmarks: 16

## Summary

- They share 16 benchmarks with published results for both. Grok-3 mini scores higher in 6 categories and Step 3 in 2 categories; 6 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where Step 3 leads 27.6 to 13.8.
- Step 3 has downloadable open weights; the other is API-only.

## Snapshot

| | Grok-3 mini | Step 3 |
|---|---|---|
| Provider | xAI | StepFun |
| Noometry Index | 41.0 | 39.9 |
| Rank | 130 | 145 |
| Context | — | — |
| Input $/M | — | — |
| Output $/M | — | — |
| Weights | Proprietary | Open |

## Coding

- Grok-3 mini: 40.4 (#126)
- Step 3: 39.5 (#144)

| Benchmark | Grok-3 mini | Step 3 |
|---|---|---|
| LMArena Coding | 1380 | 1367 |
| Aider Polyglot | 49.3% | — |
| WeirdML | 42.6% | — |

## Reasoning

- Grok-3 mini: 13.8 (#323)
- Step 3: 27.6 (#102)

| Benchmark | Grok-3 mini | Step 3 |
|---|---|---|
| Kagi LLM Benchmark | 61.3% | 62.3% |
| LMArena Hard Prompts | 1374 | 1355 |
| ARC-AGI-2 | 0.4% | — |
| ARC-AGI-1 | 16.5% | — |
| Epoch Capabilities Index | 140.35 | — |

## Math

- Grok-3 mini: 40.3 (#78)
- Step 3: 34.9 (#148)

| Benchmark | Grok-3 mini | Step 3 |
|---|---|---|
| LMArena Math | 1386 | 1366 |
| OTIS Mock AIME 2024-2025 | 77.8% | — |
| Omni-MATH | 31.8% | — |
| MATH Level 5 | 90.9% | — |
| FrontierMath (Feb 2025 set) | 5.9% | — |

## Knowledge

- Grok-3 mini: 46.3 (#75)
- Step 3: 36.2 (#161)

| Benchmark | Grok-3 mini | Step 3 |
|---|---|---|
| LMArena Expert | 1395 | 1333 |
| GPQA Diamond | 76.3% | — |
| MMLU-Pro | 79.9% | — |
| Confabulations | 10.8% | — |
| GPQA (HELM) | 67.5% | — |

## Multimodal

- Grok-3 mini: —
- Step 3: 34.9 (#81)

| Benchmark | Grok-3 mini | Step 3 |
|---|---|---|
| LMArena Vision | — | 1176 |

## Multilingual

- Grok-3 mini: 48.3 (#141)
- Step 3: 46.5 (#154)

| Benchmark | Grok-3 mini | Step 3 |
|---|---|---|
| LMArena Non-English | 1352 | 1327 |
| LMArena Chinese | 1387 | 1396 |
| LMArena German | 1350 | 1371 |
| LMArena Korean | 1336 | 1271 |
| LMArena Russian | 1352 | 1331 |
| LMArena Spanish | 1379 | 1372 |
| LMArena French | 1357 | — |
| LMArena Japanese | 1343 | — |

## Instruction Following

- Grok-3 mini: 78.7 (#8)
- Step 3: 70.6 (#159)

| Benchmark | Grok-3 mini | Step 3 |
|---|---|---|
| LMArena Instruction Following | 1356 | 1332 |
| IFEval | 95.1% | — |

## Long Context

- Grok-3 mini: 41.0 (#142)
- Step 3: 40.3 (#152)

| Benchmark | Grok-3 mini | Step 3 |
|---|---|---|
| LMArena Longer Query | 1372 | 1326 |
| Fiction.LiveBench | 66.7% | — |

## Writing & Preference

- Grok-3 mini: 52.7 (#163)
- Step 3: 54.5 (#146)

| Benchmark | Grok-3 mini | Step 3 |
|---|---|---|
| LMArena Text | 1370 | 1351 |
| LMArena Creative Writing | 1342 | 1322 |
| LMArena Multi-Turn | 1355 | 1340 |
| Short-Story Creative Writing | 73.5% | — |
| WildBench | 65.1% | — |

## FAQ

### Is Grok-3 mini better than Step 3?

Grok-3 mini is the stronger model overall, scoring 41.0 to 39.9 on the Noometry Index.

### Is Grok-3 mini or Step 3 better for coding?

They score almost the same on coding (40.4 vs 39.5); test both on your own repository before choosing.

### How many benchmarks do Grok-3 mini and Step 3 share?

16 benchmarks have published results for both models. Grok-3 mini has 35 scored results on Noometry and Step 3 has 17.