A modern, extensible playground for prompt engineering and testing against OpenRouter-hosted LLMs.
Beautiful UI, real-time web search augmentation, and seamless containerized deployment.
IntelliLLM Playground is a powerful, modern tool for prompt engineers, data scientists, and developers to create, test, and refine prompts for large language models. With a beautiful UI and rich feature set, it helps you get the most out of your LLM interactions.
- 🖥️ Modern UI: Multi-tab prompt editing, collapsible system prompt editor, and theme support
- 🤖 LLM Integration: Supports all OpenRouter models with dynamic per-prompt configuration
- 📂 Prompt Management: Import/export prompts, folder organization, and GitHub Gist integration
- 🔄 Prompt Parametrization: Create template prompts with
{{ParameterName}}syntax - ✨ Prompt Augmentation: Use AI to enhance your prompts automatically, making them more detailed and effective.
- 🧠 Web Search Augmentation: Real-time web search for all models via DuckDuckGo for up-to-date information
- 📎 File Attachments: Upload images and documents to include with your prompts
- 🐳 Dockerized: Easy to build, run, and deploy anywhere
The fastest and most reliable way to use IntelliLLM Playground is via Docker.
# 1. Clone the repo
$ git clone https://github.com/rawveg/intellillm-playground.git
$ cd intellillm-playground
# 2. Build the Docker image
$ docker build -t intellillm-playground .
# 3. Run the container (serves on port 3000)
$ docker run -p 3000:3000 -v /path/to/your/prompts:/app/prompts intellillm-playgroundTip: Mount your prompt directory (
-v /path/to/your/prompts:/app/prompts) for persistent prompt storage.
Open http://localhost:3000 in your browser.
| Layer | Technology/Notes |
|---|---|
| Frontend | Next.js, React, TailwindCSS, Radix UI, Monaco Editor |
| Backend | Next.js API routes (TypeScript) |
| LLM Access | OpenRouter API |
| Web Search | DuckDuckGo search (@pikisoft/duckduckgo-search) |
| Prompts | Markdown files with YAML frontmatter (/prompts) |
| Container | Docker (multi-stage build) |
- Feature Details
- Installation & Deployment
- Prompt Parameters
- File Attachments
- Web Search Integration
- API Reference
- Contributing
This project is licensed under the GNU Affero General Public License v3.0. See the LICENSE file for details.
- OpenRouter for LLM access
- DuckDuckGo for search integration
- @pikisoft/duckduckgo-search
- Next.js, React, TailwindCSS
Made with ❤️ for prompt engineers and LLM enthusiasts.