Model comparison
Grok 3 vs o3-mini
Grok 3 is the stronger model overall, scoring 39.9 to 36.7 on the Noometry Index.
Last verified . 31 shared benchmarks.
Summary
- They share 31 benchmarks with published results for both. Grok 3 scores higher in 7 categories and o3-mini in 2 categories; 7 gaps are clear of the uncertainty.
- The widest gap is in math, where Grok 3 leads 38.0 to 28.1.
- The biggest single-benchmark swing is ARC-AGI-1: 5.5% for Grok 3 and 34.5% for o3-mini.
Side by side
| Grok 3 | o3-mini | |
|---|---|---|
| Provider | xAI | OpenAI |
| Noometry Index | 39.9 | 36.7 |
| Released | 2025-04-09 | 2024-12-20 |
| Weights | Proprietary | Proprietary |
| Context window | — | 200K |
| Max output | — | 100K |
| Input $ / M tokens | — | $1.10 |
| Output $ / M tokens | — | $4.40 |
| Results tracked | 40 | 51 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Grok 3 leads
Grok 3: 41.9 (#115), o3-mini: 40.8 (#132)
| Benchmark | Grok 3 | o3-mini |
|---|---|---|
| Aider Polyglot | 53.3% | 60.4% |
| WeirdML | 37.2% | 43.7% |
| LMArena Coding | 1432 | 1378 |
| SciCode | — | 39.8% |
| GSO | — | 1.3% |
| LiveBench Coding | — | 82.7% |
| CadEval | — | 54% |
Agentic & Tool Use Too close to call
Grok 3: 30.5 (#76), o3-mini: 29.6 (#84)
Reasoning o3-mini leads
Grok 3: 13.7 (#333), o3-mini: 16.3 (#305)
| Benchmark | Grok 3 | o3-mini |
|---|---|---|
| ARC-AGI-2 | 0% | 3% |
| SimpleBench | 36.1% | 22.8% |
| ARC-AGI-1 | 5.5% | 34.5% |
| LMArena Hard Prompts | 1434 | 1366 |
| Epoch Capabilities Index | 138.33 | 140.34 |
| Kagi LLM Benchmark | 61.3% | — |
| CritPt | — | 0.3% |
| Chess Puzzles | — | 17% |
| LiveBench Reasoning | — | 89.6% |
| Mystery Game Puzzles | — | 7% |
| DTBench | — | 68.8% |
| LiveBench Data Analysis | — | 70.6% |
| LMCA | — | 19% |
| ForecastBench | — | 59.6 |
| LiveBench | — | 75.9% |
Math Grok 3 leads
Grok 3: 38.0 (#145), o3-mini: 28.1 (#244)
| Benchmark | Grok 3 | o3-mini |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 55.6% | 76.9% |
| LMArena Math | 1391 | 1396 |
| MATH Level 5 | 88.7% | 96.5% |
| FrontierMath (Feb 2025 set) | 3.8% | 12.4% |
| FrontierMath Tier 4 (v1) | 0% | 4.2% |
| FrontierMath (Tiers 1-3) | — | 18.6% |
| FrontierMath Tier 4 | — | 0% |
| Omni-MATH | 46.4% | — |
| LiveBench Math | — | 77.3% |
Knowledge Grok 3 leads
Grok 3: 46.2 (#82), o3-mini: 38.3 (#146)
| Benchmark | Grok 3 | o3-mini |
|---|---|---|
| GPQA Diamond | 75.8% | 77% |
| Confabulations | 14.2% | 17.9% |
| LMArena Expert | 1421 | 1364 |
| SimpleQA Verified | — | 15.3% |
| MMLU-Pro | 78.8% | — |
| Vectara Hallucination Rate | 5.8% | — |
| GPQA (HELM) | 65% | — |
Multilingual Grok 3 leads
Grok 3: 52.3 (#87), o3-mini: 45.7 (#164)
| Benchmark | Grok 3 | o3-mini |
|---|---|---|
| LMArena Non-English | 1410 | 1319 |
| LMArena Chinese | 1448 | 1379 |
| LMArena French | 1460 | 1334 |
| LMArena German | 1431 | 1303 |
| LMArena Japanese | 1387 | 1286 |
| LMArena Korean | 1373 | 1314 |
| LMArena Russian | 1416 | 1304 |
| LMArena Spanish | 1417 | 1321 |
Instruction Following Too close to call
Grok 3: 75.0 (#73), o3-mini: 75.1 (#72)
| Benchmark | Grok 3 | o3-mini |
|---|---|---|
| LMArena Instruction Following | 1409 | 1337 |
| LiveBench Instruction Following | — | 84.4% |
| IFEval | 88.4% | — |
Long Context Grok 3 leads
Grok 3: 38.7 (#192), o3-mini: 33.8 (#256)
| Benchmark | Grok 3 | o3-mini |
|---|---|---|
| Fiction.LiveBench | 58.3% | 50% |
| LMArena Longer Query | 1439 | 1343 |
Writing & Preference Grok 3 leads
Grok 3: 55.8 (#141), o3-mini: 50.3 (#182)
| Benchmark | Grok 3 | o3-mini |
|---|---|---|
| LMArena Text | 1426 | 1337 |
| LMArena Creative Writing | 1414 | 1286 |
| Short-Story Creative Writing | 76.4% | 61.7% |
| LMArena Multi-Turn | 1425 | 1320 |
| EQ-Bench Creative Writing | 1186 | — |
| WildBench | 84.9% | — |
| LiveBench Language | — | 50.7% |
Frequently asked questions
Is Grok 3 better than o3-mini?
Grok 3 is the stronger model overall, scoring 39.9 to 36.7 on the Noometry Index.
Is Grok 3 or o3-mini better for coding?
Grok 3 scores higher on coding benchmarks: 41.9 versus 40.8 in the Noometry coding category.
How many benchmarks do Grok 3 and o3-mini share?
31 benchmarks have published results for both models. Grok 3 has 40 scored results on Noometry and o3-mini has 51.