Aaahh I see it thank you! I’m not sure if this step is clear enough in the documentation though. Anyways thank you for your time!
and after that,
It took ages to download, but finally I have a couple of models downloaded and local-ai container running.
However, when I try to generate an image (texts prompts also fail) I get back a couldn’t connect error, it ranges from curl fail to connection refused or it’s a self signed certificate. Ofc depends on what adress I give for the assistant to reach the local AI instance. if I add:
http://172.19.0.2:8080 (this is the local-ai container’s ip)
OpenAI error: open /tmp/generated/images/b642151419978: no such file or directory
OpenAI error: cURL error 7: Failed to connect to localhost port 8080 after 0 ms: Couldn't connect to server (see https://curl.haxx.se/libcurl/c/libcurl-errors.html) for http://localhost:8080/v1/images/generations
I also can see a couple errors in the local-ai’s container.
rpc error: code = Unavailable desc = connection error: desc = "transport: Error while dialing: dial tcp 127.0.0.1:35663: connect: connection refused"
If I put in the address of the virtual machine where all the containers live I get:
400 Bad Request
If I enter the same, just with httpS
OpenAI error: cURL error 60: SSL certificate problem: self-signed certificate (see https://curl.haxx.se/libcurl/c/libcurl-errors.html) for https://192.168.1.18:8080/v1/images/generations
8080 is not opened up on the firewall, so I won’t check from my domain.
But based on the configs 8080 is only opened up for the master conatiner.
The hint says in in the settings
This should be the address of your LocalAI instance from the point of view of your Nextcloud server. This can be a local address with a port like http://localhost:8080 So I’m not sure why localhost not working.
Could you clear this up for me a bit?