Llama 3 was released in April 2024, with an 8 billion parameter model suitable for consumer-grade devices and a 70 billion parameter model that closed the performance gap with state-of-the-art proprietary models.

A 13 billion parameter Llama model outperformed GPT-3, which had 175 billion parameters, on most benchmarks.

Meta dropped Llama 4 on a Saturday, April 5, 2025, without warning or a coordinated press cycle.

Meta's Llama model was downloaded over a billion times within two years, making it the most downloaded AI model in history.