Pushing the frontier of cost-efficient intelligence
Grok 4 Fast is xAI's latest multimodal reasoning model, setting a new standard for cost-efficient intelligence. With a 2M token context window, it achieves a SOTA price-to-performance ratio. It's now available for free in the Grok app.
xAI's new Grok 4 Fast is built around a pretty compelling pitch.
It's fast (~190 tokens per second). It's cheap (API pricing is $0.20 / 1M input and $0.50 / 1M output tokens). And according to Artificial Analysis, this gives it a new SOTA price-to-intelligence ratio.
But the most important part is the accessibility. For a limited time, it's completely free for developers on platforms like OpenRouter. And it's also free for all users in the Grok app on web, iOS, and Android.
So, when a model is this fast, this cheap, and free for everyone to try right now... what's the reason not to?
Hi everyone!
xAI's new Grok 4 Fast is built around a pretty compelling pitch.
It's fast (~190 tokens per second). It's cheap (API pricing is $0.20 / 1M input and $0.50 / 1M output tokens). And according to Artificial Analysis, this gives it a new SOTA price-to-intelligence ratio.
But the most important part is the accessibility. For a limited time, it's completely free for developers on platforms like OpenRouter. And it's also free for all users in the Grok app on web, iOS, and Android.
So, when a model is this fast, this cheap, and free for everyone to try right now... what's the reason not to?