xAI Releases Grok Build 0.1 API for High Speed Agentic Coding

xAIxAI

· Updated

xAI launched its specialized grok-build-0.1 model in public beta via API, priced at $1 per million input tokens and $2 per million output tokens. Optimized for the low-latency loops required by autonomous agents, the model delivers over 100 tokens per second and includes native support for the Model Context Protocol.

xAI released grok-build-0.1 in public beta via its API. This model is specifically trained for agentic coding and powers the Grok Build terminal agent. It features a 256k context window and native support for the Grok Build CLI and its associated protocols.
Context window
256K tokens
Pricing (input)
$1 per million tokens
Pricing (output)
$2 per million tokens
Throughput
100+ tokens per second
Availability
xAI API, OpenRouter, Vercel

The release shifts the economics of autonomous engineering by offering coding intelligence at a fraction of the cost of general-purpose models. By optimizing for inference speed, xAI addresses the primary bottleneck in agentic loops, where tasks require dozens of internal reasoning steps and tool calls.

You can access the model through the xAI console or the Vercel AI Gateway integration. It is also available in editors like Cursor and the Kilo Code integration. Pricing is $1 per million input tokens and $2 per million output tokens, making it a competitive engine for custom agentic workflows.

View the full update on x.ai
xAI
xAI
@xai
X

grok-build-0.1 is now available via the xAI API in public beta. This is the same model that powers the Grok Build CLI and excels at agentic coding. Priced at $1/m input and $2/m output, it’s extremely cost effective, intelligent, and fast. https://t.co/2ZtqWM2QLU

1.5kretweets5.1klikes
View on X

Still wondering? A few quick answers below.

This is a specialized AI model from xAI designed specifically for agentic coding tasks like web development and debugging. Unlike general-purpose models, it is optimized for the iterative loops required by autonomous agents that plan and execute code across multiple files. It also includes native support for the Model Context Protocol to connect with external tools.

The model is positioned as a cost-effective option for developers and is priced at $1 per million input tokens and $2 per million output tokens. This pricing applies to the public beta available through the xAI API, making it significantly more economical than many frontier models for high-volume agentic workflows and tool-calling use cases.

The model is built for high-speed performance to support the low-latency requirements of autonomous agent loops. It is served at a rate of over 100 tokens per second. This speed allows agents to perform multiple internal reasoning steps and tool interactions quickly, which is critical for maintaining efficiency during complex software engineering tasks.

Beyond the direct xAI API, the model is available through several third-party providers and developer tools. You can access it via OpenRouter and the Vercel AI Gateway. It is also integrated into popular AI-native coding environments including Cursor, Hermes Agent, OpenClaw, Kilo Code, and OpenCode for use in agentic engineering workflows.

Yes, the model is currently available in public beta via the xAI API. Developers can start building with it by creating an API key through the xAI console. While it was previously accessible primarily through the Grok Build CLI for certain subscribers, this release opens the model to any developer via standard API calls.

Every HeadsUpAI update is written based on its original source and reviewed before it's published. Read our editorial standards →

See all AI news & updates from xAI →

Share this update