BenchLeader
xAIAuto-detected

Grok 3 Mini Fast

Best configuration ranks #288 of 610 on the BenchLeader Index at 50.3 ±2.3 (low reasoning effort). Last measured 1 Sept 2026.

Blended price
Output speed
First answer
Context
How it scores by categoryDashed line = average model (50). One step of 15 = one standard deviation.
  1. Overall index50
  2. Coding46
  3. Maths53
  4. Knowledge52

Benchmark results

One column per reasoning effort. Rank is among every configuration of every model on that benchmark. Hover a score for the run it came from.

Reasoning

BenchmarklowSource
GPQA Diamond (Vals)not in index73.0%#84Vals AI

Coding

BenchmarklowSource
LiveCodeBench66.3%#96Vals AI

Maths

BenchmarklowSource
AIME (Vals)70.6%#54Vals AI
MGSM90.4%#38Vals AI

Knowledge

BenchmarklowSource
MMLU-Pro80.0%#87Vals AI
LegalBench81.9%#71Vals AI
CorpFin59.5%#75Vals AI
TaxEval73.0%#55Vals AI
MedQA88.7%#52Vals AI

See also

Data as of 9 Sept 2026. Compare with another model.