xAI API is a developer platform that exposes Grok models through a single API endpoint at api.x.ai/v1. It supports text and code generation, reasoning, image and video generation and editing via the Imagine API, and realtime voice agents through speech-to-speech, text-to-speech, and speech-to-text. Developers can also search the live web and X, and upload files and collections so Grok can reason over PDFs, spreadsheets, and presentations. It works with existing SDKs, including OpenAI-compatible clients, and every Console account includes a Playground for testing models before billing. Usage is billed per token, with prepaid credits and enterprise invoicing available, and enterprise options add SSO, audit logging, data residency, and dedicated support.
How to use xAI API
1
Create an account and API key
Sign up at console.x.ai, open the API Keys page, and create a key. Keys are shown once at creation, so store them securely.
2
Point your SDK at the base URL
Set your base URL to https://api.x.ai/v1 and call a Grok model such as grok-4.7 using Python, TypeScript, an OpenAI-compatible SDK, or plain curl.
3
Test models in the Playground
Every Console account includes a Playground where you can chat with Grok and generate images and video before writing integration code.
4
Add billing and monitor usage
API usage is billed per token. Top up prepaid credits, set up billing, and monitor usage on the Console dashboard.
xAI API Features
Text and code generation with Grok models
Image and video generation and editing via the Imagine API
Realtime voice agents with speech-to-speech, text-to-speech, and speech-to-text
Live web and X search
Files and collections for reasoning over PDFs, spreadsheets, and presentations
Works with existing SDKs, including OpenAI-compatible clients
Playground included with every Console account
Enterprise controls including SSO, audit logging, and data residency
xAI API Use Cases
Build coding assistants and agent orchestration
Generate and edit images and video for campaigns
Create realtime voice agents for customer inquiries
Synthesize research with live web and X search
Extract structured data from contracts and invoices
Build internal tools and custom assistants
xAI API FAQs
Create a free account at console.x.ai, open the API Keys page, and create a key. Then point your SDK at https://api.x.ai/v1 and make your first call.
API usage is billed per token rather than free, but every Console account includes a Playground for testing the latest Grok models before adding billing.
grok-4.7 costs $2 per million input tokens and $6 per million output tokens. Full pricing for every Grok model is in the docs.
Yes. Point your existing SDK at base URL https://api.x.ai/v1 with your xAI API key, and most integrations need no other changes.
The API serves grok-4.7, grok-4.6, and grok-4.5 for text, reasoning, and coding, plus the Grok Imagine and Grok Voice model families. Older models are retired on a published schedule.
Yes. Grok models are offered in the AI catalogs of major clouds, including Microsoft, Oracle, Google, and Amazon platforms, alongside the native xAI API.
Yes. The enterprise API adds SSO, audit logging, SOC 2, HIPAA-eligible deployments with a BAA, data residency, monthly invoicing, custom rate limits, and dedicated support.
Yes. The Grok Imagine API creates and edits images and video, and the Grok Voice API handles voiceovers, dubbing, and realtime voice agents.
API pricing is usage-based and billed per token. Prices shown are in USD and exclude applicable taxes. A free Console account includes a Playground for testing models before adding billing.
Free Console account
$0USD / month
A free Console account for creating API keys and testing Grok models in the Playground before adding billing.
Create and manage API keys
Playground included with every account
Test latest Grok models before writing integration code
API usage billed separately per token
grok-4.7 text input
$2.00USD / per 1M input tokens
Usage-based pricing for text input to grok-4.7, the flagship model for coding and knowledge work with a 500k context window.
500k context window
Flagship model for coding and knowledge work
Billed per token with prepaid credits available
grok-4.7 output
$6.00USD / per 1M output tokens
Usage-based pricing for output generated by grok-4.7.
Billed per output token
Works with existing SDKs
Prepaid credits and enterprise invoicing available
Imagine API image generation
$0.04USD / per image (from)
Image generation and editing through the Imagine API, with starting prices varying by model and resolution.
Text and image to image
1K and 2K resolution options
Starting price from $0.04 per image for grok-imagine-image-2.0
Imagine API video generation
$0.02USD / per second (from)
Video generation through the Imagine API, with starting prices varying by model and resolution.
Text, image, and audio to video
480p, 720p, and 1080p options
Starting price from $0.02 per second for grok-imagine-video-1.5-lite
Voice API speech to speech
$0.08USD / per minute
Realtime speech-to-speech pricing for building voice agents.
Realtime voice agents
Billed per minute
Voice API text to speech
$15.00USD / per 1M characters
Text-to-speech pricing for voiceovers and voice agents.
Billed per 1M characters
Suitable for voiceovers and dubbing
Voice API speech to text
$0.10USD / per hour (REST)
Speech-to-text pricing, with REST at $0.10 per hour and streaming at $0.20 per hour.