Mistral: mistral-7b-instruct-v0.3:free

A high-performing, industry-standard 7.3B parameter model, with optimizations for speed and context length. An improved version of Mistral 7B Instruct v0.2, with the following changes: Extended vocabulary to 32768 Supports v3 Tokenizer Supports function calling NOTE: Support for function calling depends on the provider.