[Help] I am unable to run it successfully.

[xmenw]

TypeError: fetch failed
at Object.fetch (node:internal/deps/undici/undici:11457:11)
at process.processTicksAndRejections (node:internal/process/task_queues:95:5) {
cause: ConnectTimeoutError: Connect Timeout Error
at onConnectTimeout (ai-comic-factory/node_modules/.pnpm/[email protected]_biqbaboplfbrettd7655fr4n2y/node_modules/next/dist/compiled/undici/index.js:1:82152)
at ai-comic-factory/node_modules/.pnpm/[email protected]_biqbaboplfbrettd7655fr4n2y/node_modules/next/dist/compiled/undici/index.js:1:81644
at Immediate._onImmediate (ai-comic-factory/node_modules/.pnpm/[email protected]_biqbaboplfbrettd7655fr4n2y/node_modules/next/dist/compiled/undici/index.js:1:82034)
at process.processImmediate (node:internal/timers:476:21)
at process.callbackTrampoline (node:internal/async_hooks:130:17) {
code: 'UND_ERR_CONNECT_TIMEOUT'
}
}

I am using the INFERENCE_API method, and I have configured everything in the .env file, but every time I enter the prompt and click "Go," it gives an error. Am I using the method incorrectly? To be straightforward, I've installed all the packages and performed the standard operations. Do I still need to start Docker?

[jbilcke-hf]

Hi,

do you wish to use the default SDXL model or a custom one?

If it's a custom one, you can try first using the cloud-hosted version, to check it it works properly:

If everything works fine in the cloud app, then you can attempt to run it on your fork of the repository

usually the issue might come from a hugging face token not properly configured in the .env.local file

you can also try to use another engine (eg. replicate) to see if it changes something