v2.7.5 Windows Local and Server Model both use Llama 3 Instruct, program crash
davidngrc opened this issue · comments
davidngrc commented
I am on Windows 10 Pro.
with NVidia P40, 24G RAM
Newly download GPT4ALL v2.7.5,
only download one model: Llama 3 Instruct
steps
- I use curl to query the model in server mode, works.
- then I switch to local client, and manually load the only one download model "Llama 3 Instruct".
- after the model loaded, like at 99.9%, the program crash, and exe exit.
the problem also exists before v2.7.5
this problem also occur when use other model, or multiple models.
does not matter if you load the local model or server mode load the model first, all crash.
please help fix.
thank you
Jared Van Bortel commented
Duplicate of #2205