xAI Launches Grok Build 0.1 API for Agentic Coding

xAI has officially launched Grok Build 0.1, its fastest coding model to date, in public beta via the xAI API. Designed for agentic coding tasks, the model is optimized for workflows like web development, debugging, and multi-step code planning. It’s now live with pricing set at $1 per million tokens in and $2 per million tokens out, making it a competitive choice for developers seeking cost-effective, high-speed solutions.

The standout feature is the model’s ability to process over 100 tokens per second, powered by an architecture that combines a reasoning model with a summarizer layer to streamline outputs. This design allows Grok Build 0.1 to handle complex coding tasks, including refactoring large repositories and executing multi-step commands, all while leveraging a massive 256,000-token context window. According to xAI, this makes it ideal for autonomous software engineering workflows.

This release builds on xAI’s earlier fast-code models, but Grok Build 0.1 represents a step-change in both performance and utility. It’s already integrated with popular platforms like the Grok Build CLI, Kilo IDE extensions for VS Code and JetBrains, and tools such as Hermes Agent and OpenCode. Developers can also access the model via OpenRouter and Vercel AI Gateway, expanding its reach across ecosystems.

For practical use, xAI has shared a simple API call example to illustrate how developers can prompt the model for tasks like generating minimalist website designs. This underlines its versatility not just in coding but also in general-purpose applications requiring agentic reasoning and tool calling.

As of May 28, 2026, xAI’s Grok Build 0.1 rollout follows a string of strategic moves to solidify its position in the AI coding space. Earlier this week, the company confirmed its availability on developer-focused platforms such as Venice and announced its integration with SuperGrok and X Premium+ subscriptions, signaling an aggressive push to court enterprise and professional users.

For developers working on high-volume or complex coding tasks, Grok Build 0.1 offers a blend of speed, affordability, and advanced capabilities that could make it a go-to option in the agentic coding market. With its public beta now live, adoption rates and developer feedback will determine how effectively it can compete with rival models in this space.