xAI
Grok-3 mini benchmark results
7 results on 6 benchmarks from 4 sources.
Closed Released: Apr 9, 2025 Updated Oct 22, 2026
Also listed as: grok-3-mini-high · grok-3-mini-beta · grok-3-mini-beta_high · grok-3-mini-beta_low · Grok-3-Mini (High) · Grok-3-mini
All benchmark results
| Benchmarks | Score | Rank | Setting | Source type | Date |
|---|---|---|---|---|---|
| LMArena Text | 1364 | 201 | Overall, style control on; 15600 votes; 95% CI 1358–1369“grok-3-mini-high” | Official leaderboard | Oct 8, 2026 |
| LMArena Text | 1358 | 207 | Overall, style control on; 20937 votes; 95% CI 1353–1363“grok-3-mini-beta” | Official leaderboard | Oct 8, 2026 |
| LiveCodeBench | 62.6% | 8 | Code generation, 182 problems dated 2025-01-01…2025-04-07“Grok-3-Mini (High)” | Official leaderboard | Apr 7, 2025 |
| GPQA Diamond | 75.5% | 140 | Epoch AI run, accuracy | Independent run | May 26, 2025 |
| MMLU-Pro | 83.0% | 47 | Overall accuracy; data source: Self-Reported“Grok-3-mini” | Vendor-reported | Mar 11, 2026 |
| AIME (OTIS Mock AIME 2024–2025) | 62.2% | 144 | Epoch AI run, OTIS Mock AIME 2024–2025 | Independent run | Apr 10, 2025 |
| MATH Level 5 | 90.9% | 21 | Epoch AI run, accuracy | Independent run | Apr 10, 2025 |
FAQ
Where do these Grok-3 mini scores come from?
From Epoch AI Benchmarking Hub, LiveCodeBench leaderboard, LMArena leaderboard dataset, and MMLU-Pro leaderboard (TIGER-Lab). Each row links to the original leaderboard, dataset or report, and vendor-reported numbers are labelled.