Vendor pricing
xAI (Grok) model pricing
Every xAI (Grok) model JP prices, side by side — input, output, and cache rates per million tokens.
| Model | Input (USD / 1M) | Output (USD / 1M) | Cache read (USD / 1M) |
|---|---|---|---|
| Grok 4.5 | $2.00 | $6.00 | $0.50 |
| Grok 4.3 | $1.25 | $2.50 | $0.20 |
| Grok 4.20 Reasoning | $1.25 | $2.50 | $0.20 |
| Grok 4.20 Non-reasoning | $1.25 | $2.50 | $0.20 |
| Grok 4.20 Multi-agent | $1.25 | $2.50 | $0.20 |
| Grok Build 0.1 | $1.00 | $2.00 | $0.20 |
Notional list price. xAI (Grok) usage on a subscription is metered as a flat plan, so these per-token figures are list price, not spend.
How to read this table
All prices are USD per million tokens at published xAI API list rates (docs.x.ai/developers/pricing). Cached-input rates are model-specific on xAI — not a flat 0.1× of input — and appear in the table when the vendor publishes them.
Subscription note
On Grok Build under SuperGrok or X Premium+, these list prices are notional — what usage would cost at xAI API rates (https://docs.x.ai/developers/pricing), not subscription spend.
Common questions
- Which xAI (Grok) model is cheapest?
- Among models JP prices, Grok Build 0.1 is the lowest per-token code tier; Grok 4.3 / 4.20 are mid; Grok 4.5 is the frontier rate. The table shows exact official rates.
- Where do these xAI rates come from?
- Official xAI developer pricing at https://docs.x.ai/developers/pricing, plus each model's cached-input line on its model page. JP mirrors those list rates for notional Grok Build cost.