# Gemma 4 31B IT vs Grok 3

> Gemma 4 31B IT is the stronger model overall, scoring 43.2 to 39.6 on the Noometry Index.

- Canonical page: https://noometry.com/compare/gemma-4-31b-it-vs-grok-3
- Last updated: 2026-10-10
- Shared benchmarks: 21

## Summary

- They share 21 benchmarks with published results for both. Gemma 4 31B IT scores higher in 7 categories and Grok 3 in 1 category; 6 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where Gemma 4 31B IT leads 26.4 to 13.8.
- The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 73.3% for Gemma 4 31B IT and 55.6% for Grok 3.
- Gemma 4 31B IT has downloadable open weights; the other is API-only.

## Snapshot

| | Gemma 4 31B IT | Grok 3 |
|---|---|---|
| Provider | Google | xAI |
| Noometry Index | 43.2 | 39.6 |
| Rank | 85 | 150 |
| Context | 262K | — |
| Input $/M | $0.09 | — |
| Output $/M | $0.34 | — |
| Weights | Open | Proprietary |

## Coding

- Gemma 4 31B IT: 41.9 (#98)
- Grok 3: 41.5 (#113)

| Benchmark | Gemma 4 31B IT | Grok 3 |
|---|---|---|
| WeirdML | 52.3% | 37.2% |
| LMArena Coding | 1459 | 1432 |
| Aider Polyglot | — | 53.3% |
| LMArena WebDev | 1364 | — |
| SciCode | 43.4% | — |
| ALE-Bench | 925.5 | — |

## Agentic & Tool Use

- Gemma 4 31B IT: —
- Grok 3: 30.2 (#69)

| Benchmark | Gemma 4 31B IT | Grok 3 |
|---|---|---|
| BALROG | — | 29.5% |

## Reasoning

- Gemma 4 31B IT: 26.4 (#119)
- Grok 3: 13.8 (#322)

| Benchmark | Gemma 4 31B IT | Grok 3 |
|---|---|---|
| Kagi LLM Benchmark | 63.5% | 61.3% |
| LMArena Hard Prompts | 1448 | 1433 |
| Epoch Capabilities Index | 142.74 | 138.34 |
| ARC-AGI-2 | — | 0% |
| SimpleBench | — | 36.1% |
| NYT Connections (extended) | 70.6% | — |
| ARC-AGI-1 | — | 5.5% |
| CritPt | 1.4% | — |
| Chess Puzzles | 5% | — |
| Thematic Generalization | 53% | — |
| DTBench | 82.7% | — |
| LMCA | 39.3% | — |
| Surface Evolver Bench | 30.6% | — |

## Math

- Gemma 4 31B IT: 41.3 (#72)
- Grok 3: 36.2 (#122)

| Benchmark | Gemma 4 31B IT | Grok 3 |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 73.3% | 55.6% |
| LMArena Math | 1464 | 1390 |
| Omni-MATH | — | 46.4% |
| MATH Level 5 | — | 88.7% |
| FrontierMath (Feb 2025 set) | — | 3.8% |
| FrontierMath Tier 4 (v1) | — | 0% |

## Knowledge

- Gemma 4 31B IT: 37.9 (#141)
- Grok 3: 46.0 (#76)

| Benchmark | Gemma 4 31B IT | Grok 3 |
|---|---|---|
| GPQA Diamond | 75.8% | 75.8% |
| Vectara Hallucination Rate | 7.4% | 5.8% |
| LMArena Expert | 1465 | 1420 |
| SimpleQA Verified | 10.4% | — |
| MMLU-Pro | — | 78.8% |
| Confabulations | — | 14.2% |
| GPQA (HELM) | — | 65% |

## Multimodal

- Gemma 4 31B IT: 41.0 (#32)
- Grok 3: —

| Benchmark | Gemma 4 31B IT | Grok 3 |
|---|---|---|
| LMArena Vision | 1278 | — |
| LMArena Document | 1425 | — |

## Multilingual

- Gemma 4 31B IT: 54.0 (#52)
- Grok 3: 52.5 (#81)

| Benchmark | Gemma 4 31B IT | Grok 3 |
|---|---|---|
| LMArena Non-English | 1431 | 1411 |
| LMArena Chinese | 1478 | 1448 |
| LMArena French | 1436 | 1460 |
| LMArena Russian | 1462 | 1416 |
| LMArena Spanish | 1443 | 1417 |
| LMArena German | — | 1431 |
| LMArena Japanese | — | 1389 |
| LMArena Korean | — | 1374 |

## Instruction Following

- Gemma 4 31B IT: 75.6 (#58)
- Grok 3: 75.2 (#69)

| Benchmark | Gemma 4 31B IT | Grok 3 |
|---|---|---|
| LMArena Instruction Following | 1433 | 1409 |
| IFEval | — | 88.4% |

## Long Context

- Gemma 4 31B IT: 44.2 (#67)
- Grok 3: 38.7 (#187)

| Benchmark | Gemma 4 31B IT | Grok 3 |
|---|---|---|
| LMArena Longer Query | 1446 | 1439 |
| Fiction.LiveBench | — | 58.3% |

## Writing & Preference

- Gemma 4 31B IT: 60.7 (#92)
- Grok 3: 56.0 (#136)

| Benchmark | Gemma 4 31B IT | Grok 3 |
|---|---|---|
| LMArena Text | 1443 | 1426 |
| LMArena Creative Writing | 1416 | 1414 |
| EQ-Bench Creative Writing | 1368 | 1186 |
| LMArena Multi-Turn | 1452 | 1424 |
| Short-Story Creative Writing | — | 76.4% |
| WildBench | — | 84.9% |
| EQ-Bench 4 | 1120 | — |

## FAQ

### Is Gemma 4 31B IT better than Grok 3?

Gemma 4 31B IT is the stronger model overall, scoring 43.2 to 39.6 on the Noometry Index.

### Is Gemma 4 31B IT or Grok 3 better for coding?

They score almost the same on coding (41.9 vs 41.5); test both on your own repository before choosing.

### How many benchmarks do Gemma 4 31B IT and Grok 3 share?

21 benchmarks have published results for both models. Gemma 4 31B IT has 35 scored results on Noometry and Grok 3 has 40.