Elon Musk's AI company xAI recently launched Grok Build 0.1, an AI coding agent tool designed specifically for professional developers. This tool focuses on automatically completing complex software engineering tasks through natural language control in a local terminal, directly competing with industry-leading terminal coding products.

Recently, the third-party organization Kilo Code conducted a practical development test on Grok Build 0.1, requiring it to build a new webhook delivery service using TypeScript, Bun, and SQLite. The test results showed that the tool achieved "zero tool call failures" throughout the entire building and delivery process, demonstrating extremely stable performance.

image.png

Notably, the development cost for this complex project was only $1.65, equivalent to about 11.2 RMB. This development cost is not only far lower than competing models such as GPT-5.5 and Claude Opus 4.7, but its performance has also been praised by Elon Musk himself, who highly praised it as "an excellent value for money."

In practical details, Grok Build 0.1 demonstrated strong logic and rigor. Before finalizing any code, it first accurately extracted the signature format of Stripe, the retry behavior of GitHub, and the Standard Webhooks specification through web searches, and proactively asked the developer nine clarification questions. This deep thinking ability to clarify requirements before starting coding has become the key to its high success rate and low cost.