Model Price Index
Updated 2026-07-14 · Changelog

Grok 4.20 Non-Reasoning

xAI

xAI's Grok 4.20 in non-reasoning mode for fast, direct responses

$1.25

Input

$2.50

Output

per 1M tokens
≤200K input tokens

† Above 200K input tokens per request: $2.50 in / $5.00 out. Cache rates step at the same bracket.

Ratesper 1M tokens

Grok 4.20 Non-Reasoning pricing by input token bracket
Rate≤200K200K-1M
Standard Input$1.25$2.50
Standard Output$2.50$5.00
Cache Read$0.20$0.40

Bracket is selected by input tokens per request.

Estimate Your Monthly Cost

A request with input and output tokens, requests / month at a % cache hit rate $112.50/mo≤200K rate

Compare all models at this workload →

Per-request input tokens select the bracket. Excludes cache writes and batch discounts.

Additional Pricing

Input (>200K context)per 1M tokens$2.50
Output (>200K context)per 1M tokens$5.00

Specifications

API Model IDgrok-4.20-0309-non-reasoning
Context Window1M tokens
Modalitiestext · image
Capabilitiestool-use · streaming · json-mode · vision · code

More xAI Text / Chat Models

Grok 4.3 · Grok 4.20 Reasoning · Grok 4.20 Multi-Agent · Grok Build