xAI · model passport
Grok 3 Mini Beta
grok-3-mini-beta- Cloud only
- Proprietary licence
- Released 2025-02-19(curated)
- Overall · LMArena
- 1,358#193 of 396
- Best percentile
- 78.0LiveBench
- Cohort standing
- 52.6median 44.4
What the data says
Benchmarked on 8 of 27 tracked axes. Every figure below is the model's own published result on that board.
Uneven: 41.8 percentile points between its best and worst axis — 78 pct on LiveBench versus 36.2 pct on FrontierMath.
- LiveBench78 pct68.3%Verify →
- GPQA Diamond56.6 pct76.3%Verify →
- AIME (math)56 pct77.8%Verify →
- Instruction following53 pct1,353Verify →
- Overall51.5 pct1,358Verify →
- Hard prompts50.5 pct1,370Verify →
- Coding47.1 pct1,386Verify →
- FrontierMath36.2 pct5.9%Verify →
- WebDev ArenaNot benchmarked
- Vision ArenaNot benchmarked
- Factuality ArenaNot benchmarked
- Search ArenaNot benchmarked
- Document ArenaNot benchmarked
- Agent ArenaNot benchmarked
- SimpleQA VerifiedNot benchmarked
- ARC-AGI-2Not benchmarked
- Humanity's Last ExamNot benchmarked
- SimpleBenchNot benchmarked
- BFCL v4 (function calling)Not benchmarked
- SWE-bench Verified (live board)Not benchmarked
- Terminal-Bench 4.0Not benchmarked
- MMLU-Pro (Artificial Analysis)Not benchmarked
- LiveCodeBench (Artificial Analysis)Not benchmarked
- IFBench (Artificial Analysis)Not benchmarked
- HHEM Hallucination (Vectara)Not benchmarked
- SWE-bench VerifiedNot benchmarked
- Terminal-BenchNot benchmarked
Percentiles are the model's standing within each board's own population. Axes marked "not benchmarked" have no published result for this model — that is an absence of data, not a low score.
grok-3-mini-beta stands at 52.6 (raw mean 53.6) — measured on 8 of 27 axes.
grok-3-mini-beta stands at 52.6 versus a cohort median of 44.4 across 146 models released within 180 days — above the peer median.
On OTIS Mock AIME, grok-3-mini-beta raised the running best score by 0.8 percentage points to 77.8% on 2025-02-19.
grok-3-mini-beta ranks between #177 and #316 across 5 built-in weightings — best under "Conversation-first", worst under "All sources".
Every claim is a computed fact from the current board data, phrased by a low-temperature model call and rejected by a validation gate if any number is not in the source row. How the leaderboards work →