How to increase the timeout waiting time

[Crestina2001]

I am using o1, and it will take a lot of time to respond at times. How to increase the timeout such that it won't give an Error: Failed to fetch

[Crestina2001]

I am unfamiliar with JS. The chatGPT told me 'Error: Failed to fetch' could be thrown for various reasons, and the code does not explicitly set a timeout.

The issue may be related to the Max tokens, when I set it to 131072(suggested by the UI), it will at times throw the error, but when I set it to 131072000, it works fine. But I am still curious, as it is unlikely for the answer to exceed 131072 tokens.

[Niek]

Unfortunately streaming is not available for o1 yet. It can take really long for the model to return a reply. There is no hard timeout in the code, but we need to add some logging to show when the limit was surpassed (see: #513 (comment)).