The recent 20 million downloads could be seen as an effect of the company's Llama 3.1 update, which included a 405 billion parameter model, as well as 70 billion parameters and 8 billion parameter variants, all of which performed better on several benchmarks, such as MATH and HumanEval.
“Hosted Llama usage by token volume among our top cloud service provider partners more than doubled between May and July 2024, when we launched Llama 3.1,” Al-Dahle wrote, adding that the company’s largest LLM variant, the 405 billion parameter variant, was also gaining traction.
Furthermore, Meta has been actively trying to increase the number of partners hosting or distributing the Llama family of models. These partners include AWS, Azure, Google Cloud Platform, Databricks, Dell, Google Cloud, Groq, NVIDIA, IBM Watsonx, Scale AI, and Snowflake, among others.