- AIME'25 (cons@64; pass@1 = 82.0) [Grok 3 mini Think] 90.8
- AIME'24 (cons@64; pass@1 = 89.5) [Grok 3 mini Think] 95.8
- GPQA (cons@64; pass@1 = 80.3) [Grok 3 mini Think] 84
- LCB (Code Generation: 10/1/2024 - 2/1/2025) (cons@64; pass@1 = 74.8) [Grok 3 mini Think] 80.4
- MMLU-pro [Grok 3 mini, non-reasoning table] 78.9
- LOFT (128k) [Grok 3 mini, non-reasoning table] 83.1
- SimpleQA [Grok 3 mini, non-reasoning table] 21.7
- MMMU [Grok 3 mini, non-reasoning table] 69.4
- EgoSchema [Grok 3 mini, non-reasoning table] 74.3
Grok-3 mini
Grok-3 mini ranks first on Misguided Attention.
xAI Released 9 Apr 2025
52
Benchmarks with a score
1
Ranked first, out of five or more models
4
In the top three
xAI Apr 2025
Grok-3 mini
Closed weights
140.3
#122 of 271
Epoch Capabilities Index
- SimpleQA Verified 20.2% #41 /56
- GPQA Diamond 76.3% #97 /219
- FrontierMath 5.86% #44 /70
Where it ranks
Its best score on each benchmark, and where that ranks among every model with a score there. We rank each model once, at its best setting.
Knowledge and language
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| LLM Confabulation Leaderboard | 10.8% | #4 of 58 | Official leaderboard | |
| Fiction.LiveBench | 66.7% | #24 of 57 | Epoch AI | |
| LMArena (Elo) | 1,364 | #164 of 333 | Official leaderboard | |
| LOFT | 83.1% | #3 of 6 | Model card | |
| MMLU-Pro | 78.9% | #82 of 151 | Model card | |
| ECLeKTic | 17.1% | #20 of 35 | Official leaderboard | |
| SimpleQA | 21.7% | #38 of 59 | Model card | |
| FACTS Grounding | 75% | #22 of 34 | Official leaderboard | |
| Vectara Hallucination Leaderboard (HHEM) | 3.30% | #99 of 148 | Official leaderboard | |
| Lech Mazur Creative Writing | 7.35 | #32 of 44 | Epoch AI | |
| EQ-Bench 3 | 1,088 | #57 of 78 | Official leaderboard | |
| SimpleQA Verified | 20.2% | #41 of 56 | Official leaderboard | |
| Phare | 60% | #51 of 62 | Official leaderboard | |
| Global MMLU Lite | 67.3% | #40 of 43 | Official leaderboard |
Math and science
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| Misguided Attention | 73.7% | #1 of 45 | Official leaderboard | |
| AIME 2024 | 95.8% | #3 of 87 | Model card | |
| MATH Level 5 | 90.9% | #16 of 94 | Epoch AI | |
| AIME 2025 | 90.8% | #46 of 143 | Model card | |
| GPQA Diamond | 76.3% | #97 of 219 | Epoch AI | |
| OTIS Mock AIME 2024-2025 | 77.8% | #90 of 202 | Epoch AI | |
| PolyMath | 32.3% | #19 of 38 | Official leaderboard | |
| HMMT 2025 | 74.2% | #29 of 53 | Official leaderboard | |
| FrontierMath | 5.86% | #44 of 70 | Epoch AI | |
| BRUMO 2025 | 85% | #32 of 42 | Official leaderboard | |
| SMT 2025 | 78.8% | #31 of 40 | Official leaderboard | |
| CMIMC 2025 | 68.1% | #28 of 33 | Official leaderboard | |
| ARC-AGI | 16.5% | #79 of 92 | Official leaderboard | |
| ARC-AGI-2 | 0.42% | #85 of 95 | Official leaderboard | |
| PutnamBench | 0% | #14 of 14 | Official leaderboard |
Coding
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| LiveCodeBench | 80.4% | #11 of 73 | Model card | |
| BigCodeBench-Hard | 31.1% | #16 of 96 | Official leaderboard | |
| LiveCodeBench v6 | 66.7% | #10 of 21 | Official leaderboard | |
| Aider Polyglot | 49.3% | #27 of 54 | Epoch AI | |
| LiveCodeBench v5 | 68.4% | #11 of 21 | Official leaderboard | |
| WeirdML | 42.6% | #69 of 112 | Official leaderboard |
Agents and tools
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| Elimination Game | 6 | #6 of 56 | Official leaderboard | |
| Step Race | 3 | #16 of 59 | Official leaderboard | |
| LLM Chess | 430 | #45 of 124 | Official leaderboard | |
| LisanBench | 849.3 | #59 of 133 | Official leaderboard | |
| Berkeley Function Calling Leaderboard (BFCL) | 53% | #41 of 67 | Official leaderboard |
Professional work
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| TaxEval v2 | 73% | #54 of 129 | Official leaderboard | |
| LegalBench | 83.1% | #58 of 138 | Official leaderboard | |
| Vals Finance Agent v1 | 31.7% | #22 of 47 | Official leaderboard | |
| CorpFin v2 | 61.1% | #59 of 119 | Official leaderboard |
Vision, audio and robotics
Safety and security
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| Anthropic Red Team | 99.5% | #18 of 75 | Official leaderboard | |
| SimpleSafetyTests | 99.3% | #30 of 75 | Official leaderboard | |
| Human Pathogen Capabilities Test v2 (HPCT v2) | 51.6% | #39 of 71 | Official leaderboard | |
| HELM Safety | 90.2 | #46 of 75 | Official leaderboard | |
| AIR-Bench 2024 | 53.5% | #60 of 82 | Official leaderboard |
Other
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| HELM Capabilities | 67.9 | #24 of 60 | Official leaderboard |
xAI's models over time
xAI's models in color, against every other model on the Epoch Capabilities Index, by release date.
Its launch post
The benchmarks the lab chose to headline, with the scores it gave.