Google-Friendly News Release

Meta's Llama 2 7B Model Receives a16z Funding

The Blokes LLM GGUF Format Model Files Now Available

The Blokes LLM has released GGUF format model files for Meta's Llama 2 7B, the latest version of Meta's foundation language models. The Blokes' work was generously supported by a grant from Andreessen Horowitz (a16z).

This release includes model files for Llama-2-ko-gguf, an advanced iteration of Llama-2 with an expanded vocabulary of Korean corpus. Llama-2 is a collection of generative text models ranging in size from 7B to 70B. This release also includes model files for Ziqing Yang's Chinese Llama 2 7B.

To build your latest llama-cpp-python library with these new model files, use the following command:

 --force-reinstall --upgrade