FAQ from Mistral 7B
What is Mistral 7B?
Mistral 7B is the first large language model made free by Mistral AI for everyone. It is a powerful model adaptable to many use-cases, with natural coding abilities and 8k sequence length. It outperforms Llama 2 13B on all benchmarks and is released under Apache 2.0 licence.
How to use Mistral 7B?
To interact with Mistral 7B, you can deploy it with SkyPilot framework or use the OpenAI-compatible REST API. For deployment, you can follow the instructions provided on the website. Once deployed, you can query the model using the API.
Can I run Mistral 7B with my GPU?
Yes, Mistral 7B can be run with your gaming GPU.
How many models can I choose from?
There are two base models available: Mistral-7B-v0.1 and Mistral-7B-Instruct-v0.1.
How do I fix KeyError: 'mistral'?
You can solve this issue by upgrading your transformers version.
How do I interact with Mistral 7B using the API?
Once you have deployed the model, you can use the OpenAI-compatible REST API to query it.