Many options for running Mistral models in your terminal using LLM

The article discusses the recent developments in AI research lab, Mistral AI, which has released two powerful Large Language Models (LLMs) under an Apache 2 license, and a third larger one available via their API. The author has been testing these models using the LLM command-line tool and provides detailed instructions on how to install and run these models. The models include Mixtral 8x7B, Mistral 7B, and Mistral-medium, each with unique features and capabilities. The author also shares the impressive benchmark scores of these models, particularly highlighting the performance of Mixtral 8x7B.

The article further discusses the availability of Mistral 7B and Mixtral 8x7B under an Apache 2 license, which has led to competitive pricing from other LLM hosting providers. The author expresses concern over this trend, as it could discourage future open model releases. The author also mentions the use of LLM plugins for various providers and provides instructions on how to use them. The author concludes by inviting readers to join the #llm Discord channel to discuss their projects.

Key takeaways:

Mistral AI has released two powerful smaller Large Language Models under an Apache 2 license, and a third larger one is available via their API.
The author has been testing these models using the LLM command-line tool and provides detailed instructions on how to install and run them.
Mistral AI has also launched an early access API, La plateforme, for calling hosted versions of their models, including a new model called Mistral-medium.
There are concerns that the availability of Mistral's models under an Apache 2 license could disincentivize future open model releases due to a race to the bottom in pricing from other LLM hosting providers.

Many options for running Mistral models in your terminal using LLM

Key takeaways:

Comments (0)

Newsletter