Launching today
LLM Playground OS
Local-first playground to test compare & version LLM prompts
2 followers
Local-first playground to test compare & version LLM prompts
2 followers
🛠️ Open-source, local-first workbench to test, compare, and version LLM prompts across multiple providers—including OpenAI, Anthropic, Gemini, Groq, and Ollama. Key Features: • Compare outputs side-by-side in real time • Track prompt variations and system instructions • 100% privacy-focused: runs locally without middleman servers • Multi-model support for cloud and local models (Ollama) Designed for developers, prompt engineers, and creators building AI workflows.
Hey Product Hunt community! 👋
I built LLM-Playground-OS to solve a friction point I kept running into daily while engineering AI workflows.
The Problem
Testing prompts across different providers usually means jumping between multiple web tabs (ChatGPT, Claude, Gemini UI), copying API keys into SaaS tools with unclear privacy practices, or struggling to run side-by-side comparisons with local models like Ollama.
The Solution
LLM-Playground-OS is a 100% open-source, local-first workbench designed to give developers total ownership of their prompt testing stack:
Local-First & Private: Your API keys and prompt history stay on your local machine—no proxy servers or subscriptions.
Side-by-Side Playground: Compare output quality, latency, and response variations in real time across OpenAI, Anthropic, Gemini, Groq, and Ollama.
Prompt Versioning & History: Organize, iteration-track, and refine system prompts efficiently.
What's Next?
Since this is fully open-source, I'm actively expanding provider integrations and local testing tools. I’d love to hear your feedback, feature ideas, or any contributions on GitHub!
What does your current prompt testing setup look like?