When synthesis begins, the engine outputs clean, incremental sections directly into memory. The completed application executes safely inside this sandboxed container.
Waiting for code synthesis...Configure AI Model Provider, Model Target & API Key
Save Gemma App Builder as an offline file. Runs in any browser with zero installation.
Your key is strictly stored in sessionStorage for this browser tab only. It is wiped immediately when the tab closes.
You can select from the presets above or enter any custom model identifier.
Default: http://localhost:11434 for Ollama or http://localhost:1234 for LM Studio.
To allow in-browser fetch calls, launch Ollama with: OLLAMA_ORIGINS="*" ollama serve
No API key or local model required. Runs full 5-step incremental synthesis, code verification, buffer assembly, and live sandboxed execution for any preset app!