Artifacts

Describe an app idea and an open model generates a runnable code project with a preview.

Draw a user interface on a canvas and a vision model turns it into working HTML; uses your own API key.

Postgres running in the browser through PGlite, paired with an AI assistant for schemas, CSV imports and diagrams.

Upload a wireframe or design screenshot and get a React and Tailwind app generated from it.

Sample apps and an expression playground for the WebNN API, which runs neural networks on device hardware from the browser.

Generates a small web app from a prompt and shows the result immediately.

Upload a screenshot of a web page and get HTML that approximates it.

Describe a web page or small app and Qwen3-Coder writes the HTML, with a live preview.

Describe a small app and get generated code that previews in the page.

Generate and complete code by typing a prompt into the Code Llama playground.

Converts a sketch or figure image into TikZ code for LaTeX documents.

Writes a GitHub issue draft with SmolLM2 running locally in the browser on WebGPU.

Generate structured output such as JSON from an in-browser WebLLM model.

Spins up a mock REST, SOAP or GraphQL endpoint with rules and generated responses, and logs incoming requests.

Translates source code between more than 60 programming languages; the page says free tools work without signup.

Vector search over documents using a Postgres running in WebAssembly (PGlite) inside the browser.

Runs prompts against the language model built into Chrome, downloading it on first use.

Counts tokens for text and image attachments across several Claude models using Anthropic token-counting API and your key.

Try one JavaScript library against OpenAI, Anthropic or Gemini models by picking a provider and pasting a key.

Sends prompts with different structures to OpenAI and shows how many tokens were served from cache.