Anton Solbjørg comments

Results 209 comments of


                                            Anton Solbjørg

Creating a draft of future UI

![image](https://github.com/KillianLucas/open-interpreter/assets/4324290/e2c4c776-ff62-438d-a5c6-c1170601b54f)

Creating a draft of future UI

> Figure out how to get OI to answer to user input requests like python's input(). Do we somehow detect a delay in the output..? Is there some universal flag...

Trying to use OogaBooga as my LLM endpoint

Dont add `-l` flag. it overwrites api_base Once all the nessesary parameters in config is set, just use `interpreter` and it will load the flags from the config file

Trying to use OogaBooga as my LLM endpoint

You need to start Oobabooga with openai compatible api, se the link for more info: https://github.com/oobabooga/text-generation-webui/wiki/12-%E2%80%90-OpenAI-API

Trying to use OogaBooga as my LLM endpoint

I see, I shall work on a PR to fix this... We used to run Oobabooga with Open-interpreter using ooba before, but it was a buggy mess

Trying to use OogaBooga as my LLM endpoint

Can you try to add `--model openai/local`

Trying to use OogaBooga as my LLM endpoint

Hey, could you try this PR? Assuming you know your way around git... https://github.com/KillianLucas/open-interpreter/pull/955 After installing run interpreter with these flags: ``` -m Oobabooga/modelname # Dont know if you need...

Trying to use OogaBooga as my LLM endpoint

Install from PR: ``` pip uninstall open-interpreter pip install git pip install git+https://github.com/Notnaton/open-interpreter.git@litellm-custom-provider ``` I have updated it, now Oogabooga should work I hope

Move get_storage_path to further decouple Core from Terminal Interface

Tested on windows, updated it to upstream/main

Make the application launch faster.

When I run llama.cpp I can get 100Tok/sec on 7B model Seems interpreter have a max speed of 20tok/sec as there is little difference between 7B and a bigger model....