Grok 4.20 (non-reasoning) — pricing, context window and what it's actually good for

Grok 4.20 (non-reasoning) from xAI. 1000k context window. xAI publishes two price rows per model: one below 200k prompt tokens and one at or above. The stored headline rate is the sub-200k tier. Captured from xAI's model docs 2026-08-07 11:51 UTC.

Context window: 1,000,000 tokens
API identifier: grok-4.20-0309-non-reasoning

Pricing · per 1M tokens

Input$1.25

Output$2.50

Cached input$0.20

Cached input is ~6.3× cheaper on this model — it pays off when you send the same context again and again.

Checked against xAI model docs on 7 Aug 2026, 11:51 UTC

Tiered by PROMPT SIZE: prompts at or above 200k tokens are billed at the higher rate on input, output and cached input. Not comparable with providers that bill a flat rate across their full window.

Best for

Content Writing Customer Support Agents

Sources: xAI model docs

Not sure Grok 4.20 (non-reasoning) fits your workload?

We'll size the right model for your use case and give you a real cost estimate. One free call, no slides.

Talk to us