# Grok 4.5 vs Step 3

> Grok 4.5 is the stronger model overall, scoring 55.0 to 40.5 on the Noometry Index.

- Canonical page: https://noometry.com/compare/grok-4-5-vs-step-3
- Last updated: 2026-10-10
- Shared benchmarks: 17

## Summary

- They share 17 benchmarks with published results for both. Grok 4.5 scores higher in 9 categories and Step 3 in 0 categories; 9 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where Grok 4.5 leads 56.1 to 28.4.
- The biggest single-benchmark swing is Kagi LLM Benchmark: 83.5% for Grok 4.5 and 62.3% for Step 3.
- Step 3 has downloadable open weights; the other is API-only.

## Snapshot

| | Grok 4.5 | Step 3 |
|---|---|---|
| Provider | xAI | StepFun |
| Noometry Index | 55.0 | 40.5 |
| Rank | 25 | 149 |
| Context | 500K | — |
| Input $/M | $2 | — |
| Output $/M | $6 | — |
| Weights | Proprietary | Open |

## Coding

- Grok 4.5: 52.2 (#35)
- Step 3: 40.1 (#147)

| Benchmark | Grok 4.5 | Step 3 |
|---|---|---|
| LMArena Coding | 1474 | 1367 |
| DeepSWE | 53.8% | — |
| FrontierCode | 42.4% | — |
| LMArena WebDev | 1553 | — |
| SciCode | 54.1% | — |
| WeirdML | 46.4% | — |
| ALE-Bench | 1,309 | — |

## Agentic & Tool Use

- Grok 4.5: 44.4 (#17)
- Step 3: —

| Benchmark | Grok 4.5 | Step 3 |
|---|---|---|
| APEX-Agents | 56.2% | — |
| τ²-bench Banking | 47.9% | — |
| PostTrainBench | 23.4% | — |
| GBAEval | 65.4% | — |
| GDP.pdf | 14% | — |
| LMArena Search | 1213 | — |
| Vending-Bench 2 | 3,887 | — |

## Reasoning

- Grok 4.5: 56.1 (#25)
- Step 3: 28.4 (#105)

| Benchmark | Grok 4.5 | Step 3 |
|---|---|---|
| Kagi LLM Benchmark | 83.5% | 62.3% |
| LMArena Hard Prompts | 1462 | 1355 |
| ARC-AGI-2 | 52.6% | — |
| SimpleBench | 70% | — |
| NYT Connections (extended) | 79.9% | — |
| ARC-AGI-1 | 87.2% | — |
| CritPt | 15.4% | — |
| Chess Puzzles | 36% | — |
| DTBench | 96.5% | — |
| LMCA | 45.2% | — |
| Surface Evolver Bench | 74.4% | — |
| Epoch Capabilities Index | 153.92 | — |

## Math

- Grok 4.5: 60.9 (#35)
- Step 3: 37.6 (#148)

| Benchmark | Grok 4.5 | Step 3 |
|---|---|---|
| LMArena Math | 1459 | 1366 |
| FrontierMath (Tiers 1-3) | 57.2% | — |
| FrontierMath Tier 4 | 24.4% | — |
| OTIS Mock AIME 2024-2025 | 97.8% | — |
| ProofBench | 31% | — |

## Knowledge

- Grok 4.5: 62.3 (#24)
- Step 3: 36.8 (#164)

| Benchmark | Grok 4.5 | Step 3 |
|---|---|---|
| LMArena Expert | 1466 | 1333 |
| GPQA Diamond | 93.4% | — |
| SimpleQA Verified | 48.3% | — |

## Multimodal

- Grok 4.5: 37.6 (#72)
- Step 3: 35.5 (#86)

| Benchmark | Grok 4.5 | Step 3 |
|---|---|---|
| LMArena Vision | 1288 | 1177 |
| Blueprint-Bench 2 | 27.3% | — |
| Furniture Assembly | 22.5% | — |
| LMArena Document | 1452 | — |

## Multilingual

- Grok 4.5: 54.4 (#42)
- Step 3: 46.3 (#159)

| Benchmark | Grok 4.5 | Step 3 |
|---|---|---|
| LMArena Non-English | 1440 | 1327 |
| LMArena Chinese | 1496 | 1397 |
| LMArena German | 1446 | 1371 |
| LMArena Korean | 1404 | 1269 |
| LMArena Russian | 1448 | 1331 |
| LMArena Spanish | 1450 | 1371 |
| LMArena French | 1456 | — |
| LMArena Japanese | 1428 | — |

## Instruction Following

- Grok 4.5: 76.0 (#48)
- Step 3: 70.4 (#164)

| Benchmark | Grok 4.5 | Step 3 |
|---|---|---|
| LMArena Instruction Following | 1446 | 1332 |

## Long Context

- Grok 4.5: 44.8 (#56)
- Step 3: 40.3 (#157)

| Benchmark | Grok 4.5 | Step 3 |
|---|---|---|
| LMArena Longer Query | 1463 | 1326 |

## Writing & Preference

- Grok 4.5: 65.8 (#42)
- Step 3: 54.3 (#151)

| Benchmark | Grok 4.5 | Step 3 |
|---|---|---|
| LMArena Text | 1448 | 1350 |
| LMArena Creative Writing | 1442 | 1321 |
| LMArena Multi-Turn | 1456 | 1341 |
| EQ-Bench Creative Writing | 1579 | — |

## FAQ

### Is Grok 4.5 better than Step 3?

Grok 4.5 is the stronger model overall, scoring 55.0 to 40.5 on the Noometry Index.

### Is Grok 4.5 or Step 3 better for coding?

Grok 4.5 scores higher on coding benchmarks: 52.2 versus 40.1 in the Noometry coding category.

### How many benchmarks do Grok 4.5 and Step 3 share?

17 benchmarks have published results for both models. Grok 4.5 has 52 scored results on Noometry and Step 3 has 17.