UI for AI agents

Compare LLM responses in your UI

Evaluating models — or letting users pick a winner — needs a side-by-side surface. Compare two responses with optional blind mode and a vote bar, and give users a reusable, fillable prompt library to test with.

Components you'll use

• Model ComparisonCompare AI model responses side by side in React: optional blind mode, metadata stats, and a vote bar. Evaluate LLMs in your UI. Install via the shadcn CLI.
• Prompt TemplatesSearchable prompt template gallery in React: category filter, variable fill-in form, and onSelect. Ship a reusable prompt library. Install via the shadcn CLI.

Install

Add every component above with the shadcn CLI. You own the source after install.

pnpm dlx shadcn@latest add https://ui.vllnt.ai/r/model-comparison.json
pnpm dlx shadcn@latest add https://ui.vllnt.ai/r/prompt-templates.json

Frequently asked

How do I build an LLM comparison or eval UI?

Use Model Comparison to render two responses side by side with blind mode and a vote bar — suitable for human-preference evals. Pair it with Prompt Templates so testers can reuse curated prompts.