old torch causes version restriction due to CVE

[JaCKoP619]

LocalAI version:

latest-gpu-nvidia-cuda-12 34aece07b8ba
Environment, CPU architecture, OS, and Version:

ubuntu24.04
Describe the bug

tried to run BAAI/bge-m3 embedder, returned with:

8:28AM ERR Server error error="failed to load model with internal loader: could not load model (no success): Unexpected err=ValueError('Due to a serious vulnerability issue in `torch.load`, even with `weights_only=True`, we now require users to upgrade torch to at least v2.6 in order to use the function. This version restriction does not apply when loading files with safetensors.\\nSee the vulnerability report here https://nvd.nist.gov/vuln/detail/CVE-2025-32434'), type(err)=<class 'ValueError'>" ip=172.17.0.1 latency=14.89983365s method=POST status=500 url=/v1/embeddings

To Reproduce

-setup the container with mounted locally pulled BAAI/bge-m3 and proper .yaml file
test it via curl

Expected behavior

working embedder
Logs

Additional context