Llama 2 7B

The foundational 7-billion-parameter LLM from Meta's Llama 2 family: a high-performing, commercially-permissive base model.

Llama 2 7B is Meta AI's smallest but powerful open-release Large Language Model (LLM), featuring 7 billion parameters. It was pre-trained on a massive 2 trillion tokens and supports an extended 4,096-token context window. This model serves as the efficient entry point in the Llama 2 collection (which scales up to 70B), offering strong generative text capabilities for both research and commercial applications under its permissive license. It’s a reliable, resource-conscious option for developers building on the Llama 2 architecture.

https://huggingface.co/meta-llama/Llama-2-7b