This model is 18 months old. A newer version of it has since shipped: Grok 4.3 (2026-07-16). The details below are still accurate for this model — it just is not what we would recommend today.

Grok 3 mini

xAI🇺🇸 United States
active

Fast reasoning model from xAI. Configurable thinking effort. Strong on STEM.

Context window131K tokens
Input / 1M tokens$0.3
Output / 1M tokens$0.5

Version History

grok-3-mini-betamajor

Grok 3 mini brings fast chain-of-thought reasoning at low cost. Configurable thinking effort. Outperforms o3-mini on AIME 2025.

Benchmark Scores

Full leaderboard →
89.6%
MATH
138.0 tokens_per_sec
Speed (tok/s)