Text2vec_huggingface is not working

Description

im trying to connect weaviate to my hugging face endpoint using hf api key :

def create_weaviate_client(huggingface_key, weaviate_url, weaviate_api_key):
    headers = {"X-HuggingFace-Api-Key": huggingface_key}
    client = weaviate.connect_to_weaviate_cloud(
        cluster_url=weaviate_url,
        auth_credentials=Auth.api_key(weaviate_api_key),
        headers=headers
    )
    return client

# Import schema to Weaviate
def import_schema_to_weaviate(client, schema_data, model):
    schema_name = schema_data["database_name"]+"hi1"

    # Create the schema collection in Weaviate
    client.collections.create(
        name=schema_name,
        vectorizer_config=wvc.config.Configure.Vectorizer.text2vec_huggingface(model=model),
        vector_index_config=Configure.VectorIndex.hnsw(),  # Use the HNSW index
        properties=[
            wvc.config.Property(name="content", data_type=wvc.config.DataType.TEXT),
            wvc.config.Property(name="type", data_type=wvc.config.DataType.TEXT),
            wvc.config.Property(name="entity", data_type=wvc.config.DataType.TEXT),
            wvc.config.Property(name="description", data_type=wvc.config.DataType.TEXT)
        ]
    )
    print(f"Schema {schema_name} created successfully.")

the collelction is created successfully
but when i insert new object : for entry in schema_entries:
schema_collection.data.insert(properties=entry) , i get the following error : Error data insertion: Object was not added! Unexpected status code: 500, with response body: {‘error’: [{‘message’: ‘vectorize target vector : update vector: unmarshal error response body: Not Found’}]}.
Error data insertion: Object was not added! Unexpected status code: 500, with response body: {‘error’: [{‘message’: ‘vectorize target vector : update vector: unmarshal error response body: Not Found’}]}.

hi @4oned !!

Welcome to our community :hugs: !!

This error message seems to be an issue from the vectorizer.

What is the value of the variable model?

Here you can find more information on Huggingface available models:

Let me know if this helps!

Thanks!

Thank you so much , @DudaNogueira
The model i’m using is : BAAI/bge-base-en-v1.5 , and it worked for me before with the same code ; i guess it’s from hugging face part

Hello! I’ve been having this problem recently (past two weeks). Did you find any helpful solution? I’m doing it on java with hugging face inference.

I’ve used sentence-transformers/all-mpnet-base-v2 and haven’t found any luck. Any Suggestions Duda?

This seems to be cause by hugging face api changes: