Groq announced the exclusive launch of Meta's Llama 4 Scout and Maverick models in the Middle East. Available only on GroqCloud™, developers now have day-zero access to Meta's most advanced openly-available models.

This launch marks a significant milestone in positioning the Middle East as a hub for cutting-edge AI infrastructure, following the activation of the largest inference cluster in the region, located in Dammam. The data center, which has been live since February, is now serving Llama 4 globally.

"The integration of Llama 4 with Groq technology marks a major step forward in the Kingdom of Saudi Arabia's journey toward technological leadership," said Tareq Amin.

"We built Groq to drive the cost of compute to zero," said Jonathan Ross, CEO and Founder of Groq. "Together with our partners, we're delivering Llama 4 to the region with high-performance inference that runs faster, costs less, and doesn't compromise."

Llama 4 Now Available — Only on GroqCloud
Powered by the custom-built Groq LPU, GroqCloud gives developers instant access to Llama 4 with no tuning, no cold starts, and no trade-offs.

• Llama 4 Scout: $0.11 / M input tokens and $0.34 / M output tokens, at a blended rate of $0.13
• Llama 4 Maverick: $0.50 / M input tokens and $0.77 / M output tokens, at a blended rate of $0.53