Grok 3 Mini Fast
JSON →A fast inference variant of the Grok 3 Mini model for low-latency responses.
Specs
context window 131K tokens
max output 131K tokens
input price $0.6 / 1M tokens
output price $4 / 1M tokens
Capabilities
streamingcode-generationfunction-callingtool-useprompt-cachingreasoning
API
full doc /v1/models/grok-3-mini-fast