xAI announced on August 26, 2026, that Grok 4.6 is now available through Microsoft Foundry. The release gives organizations and developers a single platform where they can evaluate the model on their own workloads, compare it with other models and deploy managed endpoints for enterprise use.
The company describes Grok 4.6 as its latest flagship model, designed for long-running agents as well as interactive and visual work. According to xAI, the model supports a context window of 500,000 tokens, allowing individual requests to contain a large volume of input.
Developers can select from four reasoning-effort settings: low, medium, high and xhigh. These options provide configurable levels of reasoning for different tasks, although xAI’s announcement does not quantify how the settings differ in speed, cost or output quality.
Microsoft Foundry offers facilities for evaluating Grok 4.6 alongside other frontier models and running tests tailored to particular workloads, xAI said. Organizations can also deploy managed endpoints and operate them with the enterprise security and governance controls available through the platform.
xAI identifies coding agents, engineering copilots, research assistants and enterprise automation as potential applications. Grok 4.6 can be accessed through the Microsoft Foundry model catalog, while xAI points developers to its separate Microsoft Foundry integration guide for deployment instructions.
Practical context: The integration places model evaluation, workload testing and managed deployment within the same Microsoft platform, potentially simplifying adoption for organizations already using that environment. However, the announcement does not specify pricing, regional availability, comparative benchmark results or performance guarantees, so prospective users would still need to assess Grok 4.6 against their target workloads and operational requirements.
Sources
Event date: 2026-08-26. Primary source date: 2026-08-26.
