A playground closer to home.

Choose an installed model. Try a prompt, a conversation, or an embedding.
Test Ollama’s native APIs through your local frontend.

Choose your local model

Send a prompt to the model you choose.

POST

Your local response

Output from the selected Ollama model.

◎

Local intelligence, ready to explore

Choose a model and run a request. Its response will appear here.

Ready when you are◎ Ollama

⌘ Request & response inspectorView details +

NEXT REQUEST · POST /api/generate

{
  "model": "",
  "prompt": "Explain how a reading companion can help someone with weak eyesight. Use three short sentences.",
  "stream": false,
  "options": {
    "temperature": 0.7,
    "num_predict": 256
  },
  "keep_alive": "5m"
}

LAST RESPONSE

Run a request to inspect its response.

Connects to OLLAMA_API_URL (default: 127.0.0.1:11434) from the Next.js server. No Google or Sarvam key required. Only installed models are listed; inference location follows your Ollama configuration. Responses stay in page memory until reload. Requests wait for a complete response; there are no automatic retries.