search

Holmeswww / AgentKit

An intuitive LLM prompting framework for multifunctional agents, by explicitly constructing a complex "thought process" from simple natural language prompts.
Creative Commons Attribution 4.0 International
269 stars 26 forks source link

AzureOpenAI Not Working #15

Closed rhyswynn closed 2 months ago

rhyswynn commented 2 months ago

This line should be completion = client.chat.completions.create(

It is also not working when I fix this line, I am still troubleshooting, but the version I updated was working fine. Will you be making any changes to fix this, or should I just maintain my own version? Thank you

https://github.com/Holmeswww/AgentKit/blob/15000a976bbff8fc8e6480208ebaa5ec227afc6f/src/agentkit/llm_api/GPT.py#L92

rhyswynn commented 2 months ago

The other problem is because of the isInstance check, an AzureOpenAI client returns true for the isInstance check against OpenAI image

mattalford commented 2 months ago

This line should be completion = client.chat.completions.create(

It is also not working when I fix this line, I am still troubleshooting, but the version I updated was working fine. Will you be making any changes to fix this, or should I just maintain my own version? Thank you

https://github.com/Holmeswww/AgentKit/blob/15000a976bbff8fc8e6480208ebaa5ec227afc6f/src/agentkit/llm_api/GPT.py#L92

Thank you. Yes, we should be using client.chat.completions.create and we don't need to check for Azure instance. PR ready #16

Holmeswww commented 2 months ago

Merged to main, and pushed to v0.1.6.post2