xAI says demand for Grok 4 is running high on API, increased rate limits

AI company, xAI, said demand for Grok 4 on API is quite high, and developers can now experience increased default rate limits for a better development experience. According to the increased rate limits details, developers can now experience 2 million tokens per minute and 480 requests per minutes.

This new upgrade will improve their workflow and integration for a more fluid and uninterrupted model access.

xAI Increased rate limits for Grok 4 API

Last week, xAI released its most powerful AI model family, Grok 4, along with the announcement, the company also opened its access to AI developers via API. Grok 4 supports text modality with vision, image generation. Soon, it will get a specialized coding model as well as video generation.

Grok 4 on API has a 2,56,000 context window and features including functional calling, allowing you to connect the model to external tools and systems, including Cursor. Structured output allows return responses in a specific, organized format. On the other hand, reasoning model thinks before responding.

Its input cost per token for 1 million tokens is $3, the output cost per token for 1 million tokens is $15, and the cached input cost per token for 1 million tokens is $0.75.

(source)