BenchAtlas

Rankings / xAI

Grok 3 Mini Fast Low Reasoning

release date unknown

BenchAtlas Index

as of 2026-07-14
Not enough benchmark coverage for a composite score (needs ≥4 families across ≥3 categories). The per-benchmark evidence below stands on its own — sparse coverage is an evidence gap, not a low score.

Benchmark evidence

8 results
knowledge science
GPQA Diamond
implementation=vals-ai
low effort73.0 %independentVals AI
MMLU-Pro
implementation=vals-ai
low effort80.0 %independentVals AI
professional
CorpFinlow effort59.5 %independentVals AI
LegalBenchlow effort81.9 %independentVals AI
MedQAlow effort88.7 %independentVals AI
TaxEvallow effort73.0 %independentVals AI
reasoning math
AIME
implementation=vals-ai
low effort70.6 %independentVals AI
MATH-500
implementation=vals-ai
low effort70.2 %independentVals AI