⛓︎Langflow is a versatile tool for designing Language Model pipelines and has recently introduced a set of powerful new features that significantly enhance its flexibility and creative potential. Using them, users can prototype, experiment and quickly build even more powerful apps, effortlessly exploring creative ideas in the realm of language-based applications. Let’s dive into these features.
The New Prompt Template
The new prompt template allows users to create prompts with multiple variables and provides enhanced control over instructing the model for better results.
Multiple Variables
Variables can be used to define instructions, questions, context, inputs, or examples for the model and can be created with any chosen name in curly brackets, e.g. {variable_name}. They act as placeholders for parts of the text that can be easily modified or post-processed.
Custom Fields
Once inserted, these variables are immediately recognized as new fields in the prompt component. Here you have the option to define their values within the component itself or leave a field empty to be adjusted over the chat interface.
Integrating Documents
You can also use documents or output parsers as input values. By plugging them into variable handles, they’ll disable and feed that input field.
With this, users can interact with documents, webpages, or any other type of content directly from the prompt, which allows for seamless integration of external resources with the language model.
Chat Input
If working with an interactive (chat-like) flow, remember to keep one of the input variables empty to behave as the chat input.
The New Chat Interface
The updated chat interface provides a user-friendly experience and additional functionality to interact with the model and customize the prompt. The sidebar brings options that allow users to view and edit pre-defined prompt variables. This feature facilitates quick experimentation by enabling the modification of variable values right in the chat.
Notice that editing variables in the chat interface take place temporarily and won’t change their original value in the components once the chat is closed.
Displaying the Complete Prompt
To view the complete prompt in its original, structured format, simply click on the “Display Prompt” option. This feature allows you to see the prompt exactly as it entered the model.
Redefining Chat Input
In the chat interface, you can redefine which variable should be interpreted as the chat input. This gives you control over these inputs and allows for more dynamic and creative interactions.
✨ We are excited to see the community embracing Langflow as their LLM designing platform and eagerly wait to see what new inspiring projects will come out from these new features. ✨
GitHub Repo → https://bit.ly/langflow
Discord Channel → https://bit.ly/langflow-discord
