Open greygoo opened 5 days ago
This can be fixed by copying aio/cpu/text-to-text.yaml
to the models dir. It then downloads the required model and GPT-4 works in the bot as expected. Is there an easy way to add that to the preload section? If not, I'd suggest adding this step to the documentation.
LocalAI version:
quay.io/go-skynet/local-ai:v2.17.1-ffmpeg
Environment, CPU architecture, OS, and Version:
rtx4060/ryzen5700/32G
Describe the bug
When selecting the GPT-4 model in the telegram-bot example settings, and sending a prompt, the bot throws an error:
To Reproduce
Expected behavior
A reply is given to the prompt
Logs
In description
Additional context