Use huggingface-embeddings to load local embedding model

[flotos]

Is your feature request related to a problem? Please describe.

I am using the following configuration and a downloaded model as I want to deploy on an instance with no internet access

backend: huggingface-embeddings
embeddings: true
name: all-minilm
parameters:
  model: all-MiniLM-L6-v2.bin

However I get the following error at inference :

{
  "error": {
    "code": 500,
    "message": "could not load model (no success): Unexpected err=RepositoryNotFoundError('401 Client Error. (Request ID: Root=1-64e750c4-1941428a25f0e1956ea10fa1;33620328-90bf-49a5-9e52-01fa29fa39d6)\\n\\nRepository Not Found for url: https://huggingface.co/api/models/sentence-transformers/all-MiniLM-L6-v2.bin.\\nPlease make sure you specified the correct `repo_id` and `repo_type`.\\nIf you are trying to access a private or gated repo, make sure you are authenticated.\\nInvalid username or password.'), type(err)=<class 'huggingface_hub.utils._errors.RepositoryNotFoundError'>",
    "type": ""
  }
}

I indeed specified a bin file, and my other models work well so it should in theory look into the correct folder.

Describe the solution you'd like

If there is a file extension load a local model, or add a parameter for that.

[localai-bot]

⚠️⚠️⚠️⚠️⚠️

Hi! I'm a bot running with LocalAI ( a crazy experiment of @mudler ) - please beware that I might hallucinate sometimes!

but.... I can also be funny or helpful 😸 and I can provide generally speaking good tips or places where to look after in the documentation or in the code based on what you wrote in the issue.

Don't engage in conversation with me,
I don't support (yet) replying!

⚠️⚠️⚠️⚠️⚠️

The error message indicates that the requested model could not be found in the specified repository. It seems like the url parameter in the configuration is incorrect. Please double-check the spelling of the model name and ensure that it matches the actual name of the model in the Hugging Face repository. Additionally, make sure that you have the correct repo_id and repo_type specified in the configuration. If you are trying to access a private or gated repo, make sure you are authenticated with the correct username and password.
If you are certain that the URL is correct and you still receive this error, it may be possible that the model has been removed from the Hugging Face repository or is no longer available for public access. In this case, you may need to find an alternative source for the model or consider using a different model altogether.
If you would like to load a local model instead of downloading one from a repository, you can specify the local backend in your configuration and provide the path to the model file as the model parameter. For example:

backend: llama-cpp
embeddings: true
name: my-awesome-model
parameters:
  model: /path/to/my_model.bin

This will load the model from the specified local file instead of downloading it from a repository.

Sources:

• https://localai.io/features/embeddings/index.html
• https://github.com/go-skynet/LocalAI/tree/master/examples/k8sgpt/values.yaml
• https://localai.io/models/index.html

[mudler]

well - actually @localai-bot is correct here. the huggingface-embeddings backend wants a huggingface repository in the model name. See the documentation in https://localai.io/features/embeddings/index.html#huggingface-embeddings . moving to a discussion