BenchLeader
xAIAuto-detected

Grok 3 Mini Fast

Best configuration ranks #222 of 610 on the BenchLeader Index at 53.5 ±1.9 (high 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 index54
  2. Coding54
  3. Maths56
  4. Knowledge54

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

BenchmarkhighSource
GPQA Diamond (Vals)not in index79.3%#69Vals AI

Coding

BenchmarkhighSource
LiveCodeBench76.2%#75Vals AI

Maths

BenchmarkhighSource
AIME (Vals)85.0%#38Vals AI
MGSM90.4%#37Vals AI

Knowledge

BenchmarkhighSource
MMLU-Pro81.4%#78Vals AI
LegalBench83.1%#52Vals AI
CorpFin61.1%#58Vals AI
TaxEval71.1%#81Vals AI
MedQA90.1%#48Vals AI

See also

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