grok-4.6 API — per-token pricing and code examples

Checking statusPrice version 2 · in effect since Nov 6, 2026

Price per 1M tokens

InputOfficial price $2.00$0.20
Cache readOfficial price $0.50$0.02
Cache write$0.20
OutputOfficial price $6.00$0.60

Minimum per request$0.00002

Cache rates apply when the provider reports cache tokens.

90% off

About this model

grok-4.6 is served through OpenAI Chat Completions, OpenAI Responses, Anthropic Messages. You pay $0.20 per 1M input tokens and $0.60 per 1M output tokens. That is 90% below the official price of $2.00 and $6.00. Cache reads cost $0.02 per 1M tokens when the provider reports cached tokens. It is included in budget packs from $9.99 up.

Where Grok 4.6 fits

xAI positions Grok 4.6 as a frontier model "for coding, agentic tasks, and knowledge work." That is a wider brief than pure coding: besides writing and reviewing code, grok-4.6 is meant for research summaries, drafting and other office-style work where an agent reads a lot before it answers. xAI gives Grok 4.6 a 500,000-token context window and lists text plus image input with text output. Its model page does not state a maximum output length, so plan your own limit on reply size. Function calling and structured outputs are listed as supported for grok-4.6, which covers the two features most agent frameworks lean on: tool calls in a loop and responses validated against a JSON schema.

Four effort levels on Grok 4.6, including xhigh

Grok 4.6 reasons on every request and xAI does not offer a switch to turn that off. The setting you get instead is effort, and on this model xAI documents four values: low, medium, high and xhigh. xAI's reasoning guide confirms that xhigh is available on grok-4.6. When you do not send an effort value, xAI uses high.

In practice, xhigh is the level to reserve for hard problems on Grok 4.6: a migration plan across many services, a subtle concurrency bug, or a long document that has to be reconciled against another one. For chat-style help, low or medium keeps grok-4.6 responsive. Like the other xAI reasoning models, Grok 4.6 rejects presence penalty, frequency penalty and stop sequences with an error, so leave those fields out of your requests.

What it costs on Cent AI

Calls to Grok 4.6 are paid per token from your prepaid USD balance at the four rates shown in the price box on this page: input, cache read, cache write and output. Reasoning tokens from grok-4.6 are billed as output. Because xhigh lets Grok 4.6 think longer than any other level, it is also the setting most likely to raise the output share of a bill, so measure a few typical tasks at high before making xhigh your default.

xAI publishes a cached input price for Grok 4.6, and the same caching mechanism as its other models applies. Cent AI charges cache read and cache write only when xAI reports cached tokens in the usage of that request. To make hits more likely, xAI suggests a steady prompt_cache_key on the Responses API for requests that share a prefix.

Using it from your tools

grok-4.6 works on /v1/chat/completions, /v1/responses and /v1/messages at Cent AI. Cent AI does not change your parameters silently, so a value that is not valid for Grok 4.6 returns an upstream_error with the model's status code and you can fix the request. The request below is in the /v1/responses format and asks for the highest effort level that xAI documents for this model:

{
  "model": "grok-4.6",
  "reasoning": { "effort": "xhigh" },
  "prompt_cache_key": "migration-plan-v2",
  "input": "Plan the schema migration for the attached services."
}

If you reuse this payload for a different Grok model, check the effort values xAI documents for that model before you switch the model name, since an unsupported value is not removed for you and comes back as an upstream_error.

Specification

Context window
—
Input and output
Text
Endpoints
/v1/chat/completions, /v1/responses, /v1/messages
Smallest pack
$9.99

Latency and error rate per model are on the status page. status.cent-ai.dev

What a request costs

Open the full calculator

Cost per request

90% is an assumption you can change, not a promise. Cache is billed only from cache tokens the model provider reports.

See how many such requests it covers.

Estimate

grok-4.6
Fresh input99,000 tok
Cache read891,000 tok
Output10,000 tok

Per request$0.0436
At official pricesAt official prices $0.7035

Total$0.0436

An estimate at price version 2. Your bill comes from the usage the provider reports for each request.

Save $0.6598

Code

curl https://api.cent-ai.dev/v1/chat/completions \
  -H "Authorization: Bearer $CENT_AI_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model": "grok-4.6", "messages": [{"role": "user", "content": "Hello"}]}'

Set CENT_AI_KEY to a key from the dashboard. OpenAI SDKs use https://api.cent-ai.dev/v1; Anthropic SDKs use https://api.cent-ai.dev.

Price history

Version 2, in effect since Nov 6, 2026.

  1. Effective

    per 1M tokens

    Inputold $0.10 new $0.20
    Cache readold $0.01 new $0.02
    Cache writeold $0.10 new $0.20
    Outputold $0.30 new $0.60

Compare with

All models against official prices

Questions

How much does grok-4.6 cost on Cent AI?

$0.20 per 1M input tokens and $0.60 per 1M output tokens. Cache reads cost $0.02 and cache writes $0.20 per 1M tokens. Each request is rounded up once.

Is there a minimum charge?

A successful request costs at least $0.00002. A request that ends early, because the client left or the output was cut off, pays only for its tokens.

How do I pay?

Top up a prepaid USD balance with QRIS or USDT/USDC, or buy a budget pack once. Nothing renews automatically.

What if the provider does not report usage?

The request is billed from an estimate and marked estimated on the Requests page. An estimate never contains cache tokens.

Which endpoint do I call?

/v1/chat/completions, /v1/responses, /v1/messages. The OpenAI base URL is https://api.cent-ai.dev/v1.

Try grok-4.6 with one key

One prepaid balance for every model. Pay with QRIS or USDT/USDC.

Start free