twinnydotdev/twinny

Type error fetch failed t.streamResponse

Closed this issue · 4 comments

FrozzDay commented

Describe the bug
i got this when trying openai compatible api with twinny

Fetch error: TypeError: fetch failed
	at Object.fetch (node:internal/deps/undici/undici:11576:11)
	at process.processTicksAndRejections (node:internal/process/task_queues:95:5)
	at async t.streamResponse (/home/user/.vscode-oss/extensions/rjmacarthy.twinny-3.11.20-universal/out/index.js:2:135485)

Desktop (please complete the following information):

  • OS: Debian 13
  • VSCodium 1.87.2

Additional context
using other provider still result the same
(i dont use litellm proxy server)

FrozzDay commented

turns out i really need to use litellm proxy to convert openai compatible api to openai compatible api

  • 😄2
rjmacarthy commented

Hey, please could you provide more information for the request you are attempting? Perhaps the API you are calling has CORS rules in place and using LiteLLM as a proxy is helping?

FrozzDay commented

no, it has no CORS. i can use it directly on other things (like sgpt cli)

FrozzDay commented

my bad guys, i realize i should not put protocol in "Hostname" it should be domain only or ip only

  • 👍1