8 posts tagged with "llm-open-connector"

The LLM Open Connector is an API specification, closely based on the OpenAI API, that lets you create API gateways and proxy servers that connect any language model to the Einstein AI Platform. Any service that implements the API specification can be connected to Einstein Studio using the Bring Your Own LLM (BYOLLM) feature.

The LLM Open Connector opens up a world of possibilities for bespoke AI applications for Salesforce customers and independent software vendors (ISVs). Check out our recipes to learn how to bring new models to Einstein Studio!

Learn how to implement Salesforce's LLM Open Connector with the Hugging Face Serverless Inference API for LLM models that support Chat Completion. We also cover how to implement and deploy the connector as a Node.js app on Heroku.

Learn how to implement Salesforce's LLM Open Connector using Amazon Web Services (AWS). We use Amazon Bedrock to host a Llama model from Meta, a Lambda for serverless compute, and an API Gateway to expose our API.

Learn how to implement Salesforce's LLM Open Connector with the Groq platform for fast AI inference. We also cover how to deploy the connector as a Flask app on Heroku with a simple web UI for testing.

Learn how to implement Salesforce's LLM Open Connector with the SambaNova platform for fast AI inference. We also cover how to deploy the connector as a Flask app on Heroku with a simple web UI for testing.

Learn how to implement Salesforce's LLM Open Connector with the IBM watsonx platform.

Learn how to implement Salesforce's LLM Open Connector with MuleSoft Anypoint Platform using its AI Chain Connector and Inference Connector.

Learn how to implement Salesforce's LLM Open Connector using MuleSoft and the MuleSoft AI Chain (MAC) project. This recipe relies on locally hosted Ollama for quick testing. The Mule application and MAC project run in CloudHub to expose the API.