About
Llama is a family of large language models developed and released by Meta AI. The models are published under open-weight licenses and have become one of the most widely adopted foundation model families in both academic research and commercial AI development. Key releases include Llama 2 (2023), Llama 3 (April 2024), and Llama 3.1 (July 2024), with models ranging from 7B to 405B parameters. Llama 3.1 405B is the largest open-weight model from Meta and competes with leading proprietary models on major benchmarks.
Models
12 · 2023–2025
Llama 2 13BLLM4K- Llama 2 70BLLM4K
- Llama 2 7BLLM4K
- Llama 2 7B ChatLLM4K
Llama 3 8B InstructLLM8K- Llama 3.1 405BLLM · Tool-using model128K tokenów
Llama 3.1 70BLLM128K- Llama 3.1 8BLLM128K
Llama 3.2 1BLLM128K- Llama 3.2 3BLLM128K
- Llama 3.3 70BLLM · Tool-using model128K
Llama 4 MaverickLLM · Multimodal · Tool-using model1M