Anthropic
+

Anthropic

Integrate with 

Anthropic

Learn more about how Narus works with Anthropic models. Connect quickly and securely using our simple API interface. There is no technical knowledge needed, just access to the API key for Claude.

Anthropic

About Anthropic

Claude is a large language model developed by Anthropic, designed with a focus on safety and transparency. It excels in natural language processing tasks, such as text generation, summarization, and conversation, while being designed to minimize harmful outputs. Anthropic’s approach emphasizes creating AI systems that are safer and easier to guide, making Claude a popular choice for businesses and researchers who prioritize responsible AI development.

Integrating Anthropic models with Narus ensures that you and your team can get the most value out of using an AI tool.

  • Prompt optimisation helps everyone craft better prompts and get more useful outputs.
  • With the prompt library you can share highly effective prompts with the whole team.
  • Keep track of how your team is using Anthropic models and the associated costs.
  • Control who has access to Anthropic models.
  • Get tailored warnings about concerning or problematic content in user prompts.

Connect to Anthropic Models

Connecting to Anthropic Models is quick and easy.

  • Get your API key from Anthropic.
    This is usually in the settings menu for users with administrative access. 
  • Add an Anthropic model to Narus.
    In the configuration section of Narus, navigate to AI Models and select Add AI model.
  • Enter a name for the connection.
    Claude is a good starting point. Paste your API key into the form and select Save.

All your users now have access to Claude. If you want to limit access to this AI model you can do that in the User Management section.

Compatible Anthropic Models

  • Claude 3-haiku
  • Claude 3.5-sonnet

Support for Anthropic models

If you can not connect Anthropic's LLMs to Narus try these troubleshooting steps:

  • Check that your API key is correct.
  • Check the status of the LLM. If it is currently unavailable or undergoing maintenance you should try again later.

If the problem persists, contact our support team.