Grok 4 token cost & pricing

Grok 4 from xAI is a frontier reasoning model with a 256K context window and native access to real-time web information. Pricing is $3 per million input and $15 per million output — matching Claude Sonnet 4.5 exactly.

Input

per 1M tokens

Output

$15

per 1M tokens

Context window

256K

tokens

Grok 4 specifications

Vendor	xAI
Input price	$3 / 1M tokens
Output price	$15 / 1M tokens
Context window	256,000 tokens
Max output	32,768 tokens
Open weights	No
Price rank	16 of 17 (cheapest first)
Best for	reasoning, realtime

What Grok 4 costs in practice

Four common workloads, priced at list rates. Monthly figures assume 30 days.

Workload	Per request	Requests/day	Per month
Chatbot · 1K in / 500 out	$0.01	10,000	$3,150
RAG Q&A · 8K in / 400 out	$0.03	5,000	$4,500
Doc summary · 30K in / 1K out	$0.10	1,000	$3,150
Code agent · 60K in / 3K out	$0.22	500	$3,375

How many requests does $100 buy?

Assuming a 1,000-token prompt and a 500-token response per request.

9,523 requests

at $0.01 each · $10.50 per 1,000 requests

How Grok 4 pricing actually works

Identical to Sonnet 4.5 on both axes with no cached-input tier published at the same discount level, which means the decision between them should be made on capability and ecosystem rather than price.

When to use Grok 4

The differentiator is recency: tasks that depend on information newer than any training cutoff, like current-events analysis or live market commentary. For static knowledge tasks it offers no cost advantage over Sonnet.

How to cut your Grok 4 bill

Because real-time access encourages long, context-heavy prompts, cap how much web context you inject per call. Fetching ten pages when two suffice is the most common source of runaway Grok costs.

Cheaper alternatives to Grok 4

Run the numbers on your own prompt

Paste your actual text into the token counter to get an exact count instead of an estimate. Nothing is uploaded.

Open the calculator →