Something went wrong during completion. Reason: backend not found

[GNBoost0]

Hello everyone! 🙋

First of all, I'd like to thank this community and more largely the entire Github community for the incredible work you make available to the world. For me, it's wonderful to see this spirit of sharing access to common creation and development! =)
(This is my first post on Github! I'm moved! Forgive me 😜)

Now to my problem!

I installed LocalAI and Chatgpt Telegram Bot following the method described in the ReadMe of @mudler 's deposit at this address: “ https://github.com/mudler/LocalAI/tree/master/examples/telegram-bot ”

I should point out that I installed everything on my external hard drive, the aim being to be able to launch Local AI from anywhere.

Being on Windows 10, I still used different functions for copying and moving files ( copy instead of cp and move instead of mv ). I didn't use the “vim config/config.yml” command, I manually modified the config.yml file to add my Telegram bot token.

I ran the command “docker-compose --env-file config/config.env up --build“.

The images were apparently downloaded correctly, however, when I send a message to my bot on Telegram, I receive this message in response:

Something went wrong during completion. Reason: backend not found: /tmp/localai/backend_data/backend-assets/grpc/gpt4all-j {“error”:{“code”:500, “message”: “backend not found: /tmp/localai/backend_data/backend-assets/grpc/gpt4all-j”, “type”:“”}} 500 {'error': {'code': 500, 'message': 'backend not found: /tmp/localai/backend_data/backend-assets/grpc/gpt4all-j', 'type': ''}} <CIMultiDictProxy('Date': 'Sat, 28 Sep 2024 10:32:01 GMT', 'Content-Type': 'application/json', 'Content-Length': '119')>

Here the screenshot of the Telegram bot :

The model is obviously loaded correctly but not its backend.
I did a full search for a file named “gpt4all-j” but without success.

It may not be necessary, but for the sake of clarity I've installed the latest version of Docker, but I've changed the location of its AppData folder on my external hard drive.

I've attached a text file named “Log cmd.exe .txt”. It contains all the command prompt lines from installation to bot response.
Here a screenshot of last lines :

I've also attached the complete folder structure in my external hard drive containing the Docker AppData folder and the complete LocalAI Folder ( E:\LocalAI ). I also put a file containing only the structure of the “chatgpt_telegram_bot” folder ( “E:\LocalAI\LocalAI\examples\telegram-bot\chatgpt_telegram_bot” ).

I apologize for the size of the files, I wanted to be as precise as possible...

I sincerely hope that the problem is due to negligence on my part and that you'll easily find the solution! 😊

Thanks in advance! ❤️

chatgpt_telegram_bot_structure.txt
LocalAI_structure.txt
Log cmd.exe .txt

[GNBoost0]

Just a friendly reminder! 🙏

I know everyone has their own things to do, but I hope you'll take the time to help me! 💗

Edit :
Since the day of my post I see you working like a mad @mudler haha! New actions every day! 👏

Thanks for your incredible work!
I'll kindly wait and download everything again when you're done! With a bit of luck my problem will no longer exist 😜

[neoacevedo]

I come here exactly for the same.

I'm just trying LocalAI from the binary. I installed 2 models type stable diffusion. When I type in the prompt the image, a basic image I want to generate I receive the same error. Trying via curl I get the same result. The error is for both models.

The models are in another drive inside my PC, the models are stored in a hard drive and the binary is in the SSD.

{"error":{"code":500,"message":"failed to load model with internal loader: backend not found: /tmp/localai/backend_data/backend-assets/grpc/stablediffusion","type":""}}