Vercel has added xAI's Grok Build 0.1 to Vercel AI Gateway, expanding the model router's coverage for agentic coding workloads. The May 20 changelog describes Grok Build 0.1 as a beta coding model in early access and says it powers the Grok Build CLI app.

The integration is exposed through the AI SDK with the model identifier xai/grok-build-0.1. Vercel notes two limits developers need to account for: reasoning effort is not configurable, and the model does not offer a non-reasoning mode. That makes the release more of a targeted coding-model option than a general-purpose model toggle.

xAI's model documentation also lists Grok Build 0.1 under coding use cases, alongside chat, image, video, and voice model categories. Taken together, the sources position the model as a specialized coding tool rather than a broad replacement for existing chat models.

For teams already using AI Gateway, the practical change is operational. They can test Grok Build 0.1 behind the same gateway pattern used for other providers, which may simplify billing, failover, and application-level model selection. The early-access label still matters, so production use should be gated by the same evaluation and fallback checks developers would apply to any new coding agent model.