grok-build-0.1
xAI
Pricing
- Input
- $1per 1M tokens
- Output
- $2per 1M tokens
- Cache read
- $0.20per 1M tokens
Long context, a prompt above 200,000 tokens: $2 input, $0.40 cached input, $4 output per 1M tokens.
Billing
On our keys, usage of this model is billed at exactly what the model's provider charges for each call, with no markup. On your own keys, your provider bills you directly at its own rates.
Price updated October 15, 2026.

