Showcase
Real, public projects and the stacks behind them. Submit yours →
Vercel AI Chatbot
AI SDK handles streaming and multiple model providers; Next.js server actions keep it simple.
View prompt & stack
“Build a Next.js chat app with streaming responses, chat history and switchable model providers.”
Open WebUI
Ollama backend gives a private, self-hosted ChatGPT-style UI.
View prompt & stack
“A self-hosted web UI for local models via Ollama with RAG and user accounts.”
LobeChat
Plugin system and multi-provider support make it a flexible chat front end.
View prompt & stack
“A modern chat UI that supports many model providers and plugins.”
LibreChat
Unifies many AI providers in one familiar interface.
View prompt & stack
“An open-source ChatGPT clone that works with many AI providers.”
AnythingLLM
Pluggable vector stores and models for chatting with your documents.
View prompt & stack
“A desktop app to chat with your documents using any LLM and vector database.”
GPT Researcher
Planner and executor agents gather sources then write a cited report.
View prompt & stack
“An agent that researches a topic online and produces a cited report.”
AutoGPT
Pioneered autonomous goal-driven agents; now a platform for building them.
View prompt & stack
“A platform to build and run autonomous AI agents that pursue goals.”
ComfyUI
Node graph gives full control over image pipelines.
View prompt & stack
“A node-based GUI for building diffusion image generation workflows.”
Stable Diffusion WebUI
Gradio makes a feature-rich local image UI quick to extend.
View prompt & stack
“A browser UI for running Stable Diffusion locally with extensions.”
llama.cpp
Plain C++ and quantization run LLMs on ordinary hardware.
View prompt & stack
“Run LLM inference efficiently on CPUs and consumer GPUs.”
screenshot-to-code
Vision models read a screenshot and emit HTML/Tailwind or React.
View prompt & stack
“Turn a screenshot of a site into working front-end code.”
bolt.new
WebContainers run a full Node environment in the browser for the agent.
View prompt & stack
“Prompt, run, edit and deploy full-stack apps in the browser.”
Whisper
Large weakly-supervised training makes it robust to noise and accents.
View prompt & stack
“General-purpose speech recognition and translation model.”
Coqui TTS
Open models for multilingual text-to-speech and voice cloning.
View prompt & stack
“A deep learning toolkit for text-to-speech.”
Browser Use
Playwright gives agents a reliable real browser to act in.
View prompt & stack
“Make websites accessible to AI agents that click, type and navigate.”