Bugs

Don't see an option with open ai gpt4 -32k context in open models options. Even when i use my azure deployment which is gpt-4-32k, I see a limit of 4000 tokens.

Same as NodeJS

The latest openai packages is v4. See: https://github.com/openai/openai-node/discussions/217

Hi, When trying a conversation with the new GPT-1106-preview model, which is supposed to have 128k context size, I see this error: `exceeds the 8000 tokens limit of gpt-4-1106-preview. Please switch to a model with larger context size or reduce your request tokens. Looks like the model can only accept 8000 tokens maximum which is incorrect. Could you please double check?

And why doesn't it show the cost for each of the prompts?

Hi when I add a mocked response to a function it doens't appear in the chat history even though the function has been called by GPT. Am I doing something wrong? Image below of result after clicking Run. You can see the mock response has been defined for that function

hey, can you please fix this overflow on the bottom? I'm running windows 10, edge.

It seems that I get errors that looks like below for pretty much any run. --- invalid_request_error -376 is less than the minimum of 1 - 'max_tokens'

Every time I run a prompt, the User message disappears and is replaced by the Assistant message. They should both exist after running it (I like what you're doing here btw)