Update backend.py - added gpt-3.5-16k and gpt-4-32k as model options (#…

…10)

* Update backend.py - added gpt-3.5-16k and gpt-4-32k as model options

To enable chats with longer context

* Update backend.py - shortened initialization of OpenAI models

Perhaps a shorter way to initialize of OpenAI models

* Update backend.py

Re-order the models

---------

Co-authored-by: Justin Waugh <916073+bluecoconut@users.noreply.github.com>

Upgrade guidance

This fixes guidance-ai/guidance#245

ignore decode errors

Features 0618 (#7)

* readme, analytics, more packages, and cuda image

* fix cuda iamge as it should be

fix search

some ui fixes due to package updates

Add Search (#5)

* add search scaffold

* basic search

* env var clean

* search path now

* fix env

---------

Co-authored-by: Kevin Huo <kevin@approximatelabs.com>

Add Starchat-beta as a new model (#4)

* Update local model to starchat-beta

https://huggingface.co/HuggingFaceH4/starchat-beta

* Update README.md

Changed starchat model link to starchat-beta

* Adding starcoderchat-beta model

* small cleanup

---------

Co-authored-by: stoparrot <you@example.com>
Co-authored-by: Justin Waugh <916073+bluecoconut@users.noreply.github.com>

Update README.md

update demo, clean errors