LanguageGUI

LanguageGUI is an open-source Figma UI kit that addresses a specific and increasingly urgent design problem: how to translate the freeform text output of large language models into structured, interactive graphical interfaces. While most chat UIs treat LLM responses as plain text in a bubble, LanguageGUI provides over 100 components—including conversational widgets, chat bubbles, pre-built screens, and prompt boxes—that allow designers to prototype richer interactions, such as displaying tables, cards, or multi-step workflows. The kit is released under the MIT license, meaning it is free for personal and commercial use without attribution, and it comes with dark mode support out of the box. However, it is important to understand what LanguageGUI is not: it is a design resource for Figma, not a code library. The team behind it—Tonki Labs, with major contributions from Mauro Sicard and Miguel Joya—has experimented with formatting LLM responses into JSON to render widgets in real time, but they have paused that effort due to the custom implementation required per developer. This means that while LanguageGUI can accelerate the design phase, the gap between a Figma prototype and a working application remains significant. For UI/UX designers, the kit offers a head start on LLM-specific patterns like chat sidebars, multi-prompt workflows, and structured output widgets that go beyond simple text. For developers, it serves as a clear design specification and a conversation starter, but they should not expect turnkey code generation. The 30+ pre-built screens and 20+ chat bubbles cover common use cases—from a standard chatbot to a data-rich assistant that returns tables or charts—and the 10+ conversational widgets are particularly valuable for teams building interfaces that need to display structured information alongside conversational flow. The multi-prompt workflow screens are a standout, as they address a pattern that is increasingly common in complex LLM applications: guiding users through a sequence of prompts or steps. Yet, the kit's limitations are real. It is Figma-only, with no code export or runtime rendering; the community-driven development means that future code library support is uncertain; and the sheer number of components can be overwhelming without a clear design system strategy. Teams should evaluate LanguageGUI as a prototyping accelerator and a shared vocabulary for LLM interface design, but they must plan for the translation work needed to bring those designs to production. For those who need a free, open-source starting point for conversational AI UIs, LanguageGUI is a solid foundation—just be prepared to build the bridge from design to code yourself.