# Grok 3 vs o1-mini

> Grok 3 is the stronger model overall, scoring 39.6 to 33.7 on the Noometry Index.

- Canonical page: https://noometry.com/compare/grok-3-vs-o1-mini
- Last updated: 2026-10-10
- Shared benchmarks: 29

## Summary

- They share 29 benchmarks with published results for both. Grok 3 scores higher in 8 categories and o1-mini in 1 category; 9 gaps are clear of the uncertainty.
- The widest gap is in knowledge, where Grok 3 leads 46.0 to 34.8.
- The biggest single-benchmark swing is Aider Polyglot: 53.3% for Grok 3 and 32.9% for o1-mini.

## Snapshot

| | Grok 3 | o1-mini |
|---|---|---|
| Provider | xAI | OpenAI |
| Noometry Index | 39.6 | 33.7 |
| Rank | 150 | 230 |
| Context | — | — |
| Input $/M | — | — |
| Output $/M | — | — |
| Weights | Proprietary | Proprietary |

## Coding

- Grok 3: 41.5 (#113)
- o1-mini: 35.1 (#218)

| Benchmark | Grok 3 | o1-mini |
|---|---|---|
| Aider Polyglot | 53.3% | 32.9% |
| WeirdML | 37.2% | 36.3% |
| LMArena Coding | 1432 | 1362 |
| LiveBench Coding | — | 48% |
| HumanEval+ | — | 89% |
| MBPP+ | — | 78.8% |

## Agentic & Tool Use

- Grok 3: 30.2 (#69)
- o1-mini: 24.3 (#112)

| Benchmark | Grok 3 | o1-mini |
|---|---|---|
| Cybench | — | 10% |
| BALROG | 29.5% | — |

## Reasoning

- Grok 3: 13.8 (#322)
- o1-mini: 8.9 (#340)

| Benchmark | Grok 3 | o1-mini |
|---|---|---|
| ARC-AGI-2 | 0% | 0.8% |
| SimpleBench | 36.1% | 18.1% |
| ARC-AGI-1 | 5.5% | 14% |
| LMArena Hard Prompts | 1433 | 1333 |
| Epoch Capabilities Index | 138.34 | 135.82 |
| Kagi LLM Benchmark | 61.3% | — |
| LiveBench Reasoning | — | 72.3% |
| LiveBench Data Analysis | — | 57.9% |
| LiveBench | — | 57.8% |

## Math

- Grok 3: 36.2 (#122)
- o1-mini: 33.7 (#168)

| Benchmark | Grok 3 | o1-mini |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 55.6% | 46.9% |
| LMArena Math | 1390 | 1358 |
| MATH Level 5 | 88.7% | 89.2% |
| FrontierMath (Feb 2025 set) | 3.8% | 1.7% |
| Omni-MATH | 46.4% | — |
| LiveBench Math | — | 62% |
| FrontierMath Tier 4 (v1) | 0% | — |

## Knowledge

- Grok 3: 46.0 (#76)
- o1-mini: 34.8 (#180)

| Benchmark | Grok 3 | o1-mini |
|---|---|---|
| GPQA Diamond | 75.8% | 62.4% |
| Confabulations | 14.2% | 18.6% |
| LMArena Expert | 1420 | 1316 |
| MMLU-Pro | 78.8% | — |
| Vectara Hallucination Rate | 5.8% | — |
| GPQA (HELM) | 65% | — |

## Multilingual

- Grok 3: 52.5 (#81)
- o1-mini: 43.8 (#177)

| Benchmark | Grok 3 | o1-mini |
|---|---|---|
| LMArena Non-English | 1411 | 1289 |
| LMArena Chinese | 1448 | 1314 |
| LMArena French | 1460 | 1292 |
| LMArena German | 1431 | 1278 |
| LMArena Japanese | 1389 | 1245 |
| LMArena Korean | 1374 | 1223 |
| LMArena Russian | 1416 | 1283 |
| LMArena Spanish | 1417 | 1302 |

## Instruction Following

- Grok 3: 75.2 (#69)
- o1-mini: 67.0 (#200)

| Benchmark | Grok 3 | o1-mini |
|---|---|---|
| LMArena Instruction Following | 1409 | 1304 |
| LiveBench Instruction Following | — | 65.4% |
| IFEval | 88.4% | — |

## Long Context

- Grok 3: 38.7 (#187)
- o1-mini: 40.1 (#156)

| Benchmark | Grok 3 | o1-mini |
|---|---|---|
| LMArena Longer Query | 1439 | 1321 |
| Fiction.LiveBench | 58.3% | — |

## Writing & Preference

- Grok 3: 56.0 (#136)
- o1-mini: 48.6 (#196)

| Benchmark | Grok 3 | o1-mini |
|---|---|---|
| LMArena Text | 1426 | 1317 |
| LMArena Creative Writing | 1414 | 1244 |
| Short-Story Creative Writing | 76.4% | 64.9% |
| LMArena Multi-Turn | 1424 | 1314 |
| EQ-Bench Creative Writing | 1186 | — |
| WildBench | 84.9% | — |
| LiveBench Language | — | 40.9% |

## FAQ

### Is Grok 3 better than o1-mini?

Grok 3 is the stronger model overall, scoring 39.6 to 33.7 on the Noometry Index.

### Is Grok 3 or o1-mini better for coding?

Grok 3 scores higher on coding benchmarks: 41.5 versus 35.1 in the Noometry coding category.

### How many benchmarks do Grok 3 and o1-mini share?

29 benchmarks have published results for both models. Grok 3 has 40 scored results on Noometry and o1-mini has 39.