# Granite 4.2 3b vs Grok 4.5

> Grok 4.5 is the stronger model overall, scoring 55.0 to 39.0 on the Noometry Index.

- Canonical page: https://noometry.com/compare/granite-4-2-3b-vs-grok-4-5
- Last updated: 2026-10-10
- Shared benchmarks: 11

## Summary

- They share 11 benchmarks with published results for both. Granite 4.2 3b scores higher in 0 categories and Grok 4.5 in 7 categories; 7 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where Grok 4.5 leads 56.0 to 25.0.
- Granite 4.2 3b has downloadable open weights; the other is API-only.

## Snapshot

| | Granite 4.2 3b | Grok 4.5 |
|---|---|---|
| Provider | IBM | xAI |
| Noometry Index | 39.0 | 55.0 |
| Rank | 163 | 22 |
| Context | — | 500K |
| Input $/M | — | $2 |
| Output $/M | — | $6 |
| Weights | Open | Proprietary |

## Coding

- Granite 4.2 3b: 39.4 (#146)
- Grok 4.5: 51.7 (#34)

| Benchmark | Granite 4.2 3b | Grok 4.5 |
|---|---|---|
| LMArena Coding | 1363 | 1473 |
| DeepSWE | — | 53.8% |
| FrontierCode | — | 42.4% |
| LMArena WebDev | — | 1552 |
| SciCode | — | 54.1% |
| WeirdML | — | 46.4% |
| ALE-Bench | — | 1,309 |

## Agentic & Tool Use

- Granite 4.2 3b: —
- Grok 4.5: 44.4 (#15)

| Benchmark | Granite 4.2 3b | Grok 4.5 |
|---|---|---|
| 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

- Granite 4.2 3b: 25.0 (#136)
- Grok 4.5: 56.0 (#23)

| Benchmark | Granite 4.2 3b | Grok 4.5 |
|---|---|---|
| LMArena Hard Prompts | 1308 | 1462 |
| ARC-AGI-2 | — | 52.6% |
| SimpleBench | — | 70% |
| Kagi LLM Benchmark | — | 83.5% |
| 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 | — | 154.02 |

## Math

- Granite 4.2 3b: —
- Grok 4.5: 60.7 (#29)

| Benchmark | Granite 4.2 3b | Grok 4.5 |
|---|---|---|
| FrontierMath (Tiers 1-3) | — | 57.2% |
| FrontierMath Tier 4 | — | 24.4% |
| OTIS Mock AIME 2024-2025 | — | 97.8% |
| ProofBench | — | 31% |
| LMArena Math | — | 1457 |

## Knowledge

- Granite 4.2 3b: 35.6 (#169)
- Grok 4.5: 62.3 (#21)

| Benchmark | Granite 4.2 3b | Grok 4.5 |
|---|---|---|
| LMArena Expert | 1313 | 1466 |
| GPQA Diamond | — | 93.4% |
| SimpleQA Verified | — | 48.3% |

## Multimodal

- Granite 4.2 3b: —
- Grok 4.5: 37.8 (#63)

| Benchmark | Granite 4.2 3b | Grok 4.5 |
|---|---|---|
| LMArena Vision | — | 1287 |
| Blueprint-Bench 2 | — | 27.3% |
| Furniture Assembly | — | 22.5% |
| LMArena Document | — | 1452 |

## Multilingual

- Granite 4.2 3b: 42.4 (#193)
- Grok 4.5: 54.5 (#43)

| Benchmark | Granite 4.2 3b | Grok 4.5 |
|---|---|---|
| LMArena Non-English | 1268 | 1439 |
| LMArena Chinese | 1268 | 1497 |
| LMArena Russian | 1249 | 1447 |
| LMArena French | — | 1458 |
| LMArena German | — | 1448 |
| LMArena Japanese | — | 1425 |
| LMArena Korean | — | 1402 |
| LMArena Spanish | — | 1452 |

## Instruction Following

- Granite 4.2 3b: 67.2 (#195)
- Grok 4.5: 76.3 (#45)

| Benchmark | Granite 4.2 3b | Grok 4.5 |
|---|---|---|
| LMArena Instruction Following | 1271 | 1446 |

## Long Context

- Granite 4.2 3b: 39.3 (#179)
- Grok 4.5: 44.8 (#54)

| Benchmark | Granite 4.2 3b | Grok 4.5 |
|---|---|---|
| LMArena Longer Query | 1293 | 1463 |

## Writing & Preference

- Granite 4.2 3b: 47.9 (#202)
- Grok 4.5: 65.9 (#40)

| Benchmark | Granite 4.2 3b | Grok 4.5 |
|---|---|---|
| LMArena Text | 1295 | 1448 |
| LMArena Creative Writing | 1215 | 1442 |
| LMArena Multi-Turn | 1293 | 1457 |
| EQ-Bench Creative Writing | — | 1579 |

## FAQ

### Is Granite 4.2 3b better than Grok 4.5?

Grok 4.5 is the stronger model overall, scoring 55.0 to 39.0 on the Noometry Index.

### Is Granite 4.2 3b or Grok 4.5 better for coding?

Grok 4.5 scores higher on coding benchmarks: 51.7 versus 39.4 in the Noometry coding category.

### How many benchmarks do Granite 4.2 3b and Grok 4.5 share?

11 benchmarks have published results for both models. Granite 4.2 3b has 11 scored results on Noometry and Grok 4.5 has 52.