Skip to main content

Anthropic Generative AI with Weaviate

Weaviate's integration with Anthropic's APIs allows you to access their models' capabilities directly from Weaviate.

Configure a Weaviate collection to use an Anthropic generative AI model, and Weaviate will perform retrieval augmented generation (RAG) using the specified model and your Anthropic API key.

More specifically, Weaviate will perform a search, retrieve the most relevant objects, and then pass them to the Anthropic generative model to generate outputs.

RAG integration illustration

Requirements​

Weaviate configuration​

Your Weaviate instance must be configured with the Anthropic generative AI integration (generative-anthropic) module.

For Weaviate Cloud (WCD) users

This integration is enabled by default on Weaviate Cloud (WCD) serverless instances.

For self-hosted users

API credentials​

You must provide a valid Anthropic API key to Weaviate for this integration. Go to Anthropic to sign up and obtain an API key.

Provide the API key to Weaviate using one of the following methods:

  • Set the ANTHROPIC_APIKEY environment variable that is available to Weaviate.
  • Provide the API key at runtime, as shown in the examples below.
import weaviate
from weaviate.classes.init import Auth
import os

# Recommended: save sensitive data as environment variables
anthropic_key = os.getenv("ANTHROPIC_APIKEY")
headers = {
    "X-Anthropic-Api-Key": anthropic_key,
    "X-Anthropic-Baseurl": "https://api.anthropic.com",  # Optional; for providing a custom base URL
}

client = weaviate.connect_to_weaviate_cloud(
    cluster_url=weaviate_url,                       # `weaviate_url`: your Weaviate URL
    auth_credentials=Auth.api_key(weaviate_key),      # `weaviate_key`: your Weaviate API key
    headers=headers
)

# Work with Weaviate

client.close()

Configure collection​

Generative model integration mutability

A collection's generative model integration configuration is mutable from v1.25.23, v1.26.8 and v1.27.1. See this section for details on how to update the collection configuration.

Configure a Weaviate index as follows to use an Anthropic generative model:

from weaviate.classes.config import Configure

client.collections.create(
    "DemoCollection",
    generative_config=Configure.Generative.anthropic()
    # Additional parameters not shown
)

Select a model​

You can specify one of the available models for Weaviate to use, as shown in the following configuration example:

from weaviate.classes.config import Configure

client.collections.create(
    "DemoCollection",
    generative_config=Configure.Generative.anthropic(
        model="claude-3-opus-20240229"
    )
    # Additional parameters not shown
)

You can specify one of the available models for Weaviate to use. The default model is used if no model is specified.

Generative parameters​

Configure the following generative parameters to customize the model behavior.

from weaviate.classes.config import Configure

client.collections.create(
    "DemoCollection",
    generative_config=Configure.Generative.anthropic(
        # # These parameters are optional
        # base_url="https://api.anthropic.com",
        # model="claude-3-opus-20240229",
        # max_tokens=512,
        # temperature=0.7,
        # stop_sequences=["\n\n"],
        # top_p=0.9,
        # top_k=5,
    )
    # Additional parameters not shown
)

For further details on model parameters, see the Anthropic API documentation.

Retrieval augmented generation​

After configuring the generative AI integration, perform RAG operations, either with the single prompt or grouped task method.

Single prompt​

Single prompt RAG integration generates individual outputs per search result

To generate text for each object in the search results, use the single prompt method.

The example below generates outputs for each of the n search results, where n is specified by the limit parameter.

When creating a single prompt query, use braces {} to interpolate the object properties you want Weaviate to pass on to the language model. For example, to pass on the object's title property, include {title} in the query.

collection = client.collections.get("DemoCollection")

response = collection.generate.near_text(
    query="A holiday film",  # The model provider integration will automatically vectorize the query
    single_prompt="Translate this into French: {title}",
    limit=2
)

for obj in response.objects:
    print(obj.properties["title"])
    print(f"Generated output: {obj.generated}")  # Note that the generated output is per object

Grouped task​

Grouped task RAG integration generates one output for the set of search results

To generate one text for the entire set of search results, use the grouped task method.

In other words, when you have n search results, the generative model generates one output for the entire group.

collection = client.collections.get("DemoCollection")

response = collection.generate.near_text(
    query="A holiday film",  # The model provider integration will automatically vectorize the query
    grouped_task="Write a fun tweet to promote readers to check out these films.",
    limit=2
)

print(f"Generated output: {response.generated}")  # Note that the generated output is per query
for obj in response.objects:
    print(obj.properties["title"])

References​

Maximum output tokens​

Use the maxTokens parameter to set the maximum number of output tokens for the Anthropic Generative AI models. This parameter is separate from the maximum allowable input tokens, also called a "context window".

For most models, the default maxTokens value is 4096, which is the maximum, and the input (context window) size is 200,000. The specific allowable values may vary between models. Refer to the Anthropic documentation for the latest information.

Base URL​

Note that for Anthropic, you can provide a custom base URL for the API endpoint. This is useful for users who have a dedicated API endpoint, or is behind a proxy.

The custom base URL can be provided via the collection configuration as shown above, or in the header of the request. To provide it in the header, instantiate the Weaviate client using the X-Anthropic-Baseurl key and the custom base URL as the value.

The default base URL is https://api.anthropic.com.

Available models​

Any model available in the Anthropic API can be used with Weaviate. As of July 2024, the following models are available:

  • claude-3-5-sonnet-20240620 (default)
  • claude-3-opus-20240229
  • claude-3-sonnet-20240229
  • claude-3-haiku-20240307

See the Anthropic API documentation for the most up-to-date list of available models.

Further resources​

Code examples​

Once the integrations are configured at the collection, the data management and search operations in Weaviate work identically to any other collection. See the following model-agnostic examples:

  • The how-to: manage data guides show how to perform data operations (i.e. create, update, delete).
  • The how-to: search guides show how to perform search operations (i.e. vector, keyword, hybrid) as well as retrieval augmented generation.

References​

Questions and feedback​

If you have any questions or feedback, let us know in the user forum.