How much does Grok 4.6 cost on the Google Enterprise Agent Platform?

xAI lists three rates for Grok 4.6 on Google's platform: $2 per 1 million input tokens, $6 per 1 million output tokens, and $0.50 per 1 million cached input tokens. The cached rate is a quarter of the standard input price, so prompts that repeat the same long prefix cost far less to send again.

Where do I find Grok 4.6 inside Google Cloud?

Grok 4.6 sits in Model Garden under the xAI publisher on the Google Enterprise Agent Platform. xAI's post points developers at the Grok 4.6 model card in Model Garden, and Google publishes matching partner-model documentation under its Gemini Enterprise Agent Platform docs.

Does Grok 4.6 keep its full context window and reasoning settings here?

Yes. The listing gives Grok 4.6 a 500K-token context window and the same four configurable reasoning efforts xAI ships elsewhere: low, medium, high and xhigh. Teams can dial effort down for cheap, fast calls and up for the slower steps of a long task.

What kind of work does xAI say Grok 4.6 is built for?

xAI describes Grok 4.6 as its latest flagship model, built for long-running agents and for ambitious interactive and visual work. That framing matches the 500K context window, which gives an agent room to hold a long task history instead of restarting from a summary.

xAI · 2026-08-21 · major

Grok 4.6 on Google's agent platform — xAI's flagship arrives in Model Garden

Grok 4.6 is now available on the Google Enterprise Agent Platform through Model Garden. xAI's flagship keeps its 500K-token context window and four reasoning levels, at $2 per million input tokens and $6 per million output.

xAI announcement card for Grok 4.6 on the Google Enterprise Agent Platform

Grok 4.6 is listed in Google's Model Garden with a 500K context window and $2 / $6 per million tokens.

Quick facts

Maker	xAI
Platform	Google Enterprise Agent Platform (Model Garden)
Context window	500K tokens
Reasoning efforts	low, medium, high, xhigh
Price (input)	$2 / 1M tokens
Price (cached input)	$0.50 / 1M tokens
Price (output)	$6 / 1M tokens

Pricing

Input	$2.00 / 1M tokens
Cached input	$0.50 / 1M tokens
Output	$6.00 / 1M tokens

source ↗

What is it?

Google's Enterprise Agent Platform now serves Grok 4.6, which xAI announced on August 21, 2026. The model shows up in Model Garden under the xAI publisher, so Google Cloud customers can call xAI's flagship from the same platform they already use for Gemini and other partner models. xAI published the token prices with the listing.

How does it work?

A developer picks the Grok 4.6 model card in Model Garden, then sets a reasoning effort — low, medium, high or xhigh — for each call. Requests run inside a 500K-token context window. Billing splits three ways: standard input, output, and a cheaper cached-input rate for prompt content that is sent again.

Why does it matter?

Teams standardised on Google Cloud no longer have to open a separate xAI account to reach Grok 4.6. Buying through Model Garden keeps the model inside the identity, quota and billing setup a Google Cloud org already runs, which is usually the blocker for regulated buyers rather than the model quality itself. The published cached-input rate also makes long-context agent loops easier to budget.

Who is it for?

Google Cloud teams building long-running agents

Frequently asked questions

How much does Grok 4.6 cost on the Google Enterprise Agent Platform?: xAI lists three rates for Grok 4.6 on Google's platform: $2 per 1 million input tokens, $6 per 1 million output tokens, and $0.50 per 1 million cached input tokens. The cached rate is a quarter of the standard input price, so prompts that repeat the same long prefix cost far less to send again.
Where do I find Grok 4.6 inside Google Cloud?: Grok 4.6 sits in Model Garden under the xAI publisher on the Google Enterprise Agent Platform. xAI's post points developers at the Grok 4.6 model card in Model Garden, and Google publishes matching partner-model documentation under its Gemini Enterprise Agent Platform docs.
Does Grok 4.6 keep its full context window and reasoning settings here?: Yes. The listing gives Grok 4.6 a 500K-token context window and the same four configurable reasoning efforts xAI ships elsewhere: low, medium, high and xhigh. Teams can dial effort down for cheap, fast calls and up for the slower steps of a long task.
What kind of work does xAI say Grok 4.6 is built for?: xAI describes Grok 4.6 as its latest flagship model, built for long-running agents and for ambitious interactive and visual work. That framing matches the 500K context window, which gives an agent room to hold a long task history instead of restarting from a summary.

Try it

Open the Grok 4.6 model card in Google Cloud Model Garden under the xAI publisher