Gorilla models for API calls

[slobentanzer]

Model name

Gorilla: https://gorilla.cs.berkeley.edu

Parameters

Not sure testing the API-tuned model in the same benchmark. Maybe develop an API benchmark, maybe related to the function calling development by @Tatan47

Source

https://inference.readthedocs.io/en/latest/models/builtin/llm/gorilla-openfunctions-v2.html#models-llm-gorilla-openfunctions-v2
(there is also v1, not sure how they differ; maybe just include both?)