Skip to main content

LLMExecutor

Overview​

LLMExecutor, which stands for Large Language Model Executor, is responsible for passing prompts to the Large Language Model interface for processing.

Features​

Prompt template​

LLMExecutor provides prompt templates, allowing you to embed various types of messages, including those from the system, user, and assistant.

images of prompt template

Prompt Params​

Within messages, you can define a variable using double curly braces ( {{}} ). Then, by setting Params during runtime, you can assign values to these variables, thereby enabling the dynamic capability of prompts.

images of prompt params

Service Provider​

You need to configure a service provider to communicate with the Large Language Model interface. Babel includes a built-in service provider that offers a certain amount of free usage every month for each user. This can be utilized to explore the capabilities of the LLMExecutor Element.

Currently, the built-in service provider only supports the gpt-3.5-turbo-16k model.

images of LLMExecutor Element - built-in service provider

For serious applications, it is recommended to use the OPENAI service provider.

images of LLMExecutor Element - openai service provider

Access​

Access in the Prompt test​

Once you have finished crafting your prompt, you can test it.

images of LLMExecutor Element - prompt test

Access in the LLMExecutor SDK​

If you need to access LLMExecutor Element within your code, please refer to the LLMExecutor SDK section.