HPP Router

Playground

Pick a model, check the API docs, and test LLM calls — use your API key right in the browser.

Need an API key? Get one from your account.

Test LLM Inference

Select Model
Image (optional)PNG/JPEG/GIF/WebP · max 5

Your Usage

Enter your API key above to view usage statistics.

API Endpoints

OpenAI Compatible
endpoint
http://localhost:8000/llm/v1/chat/completions

Use provider/model or hpprouter/auto for smart routing.

Native Anthropic (Claude Code)
endpoint
http://localhost:8000/v1/messages
Example — chat (curl)
curl http://localhost:8000/llm/v1/chat/completions \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "openai/gpt-4o",
    "messages": [{"role": "user", "content": "Hello!"}],
    "max_tokens": 256
  }'
Example — vision / multimodal (curl)
# Encode image: BASE64_IMAGE_DATA=$(base64 -i photo.png | tr -d '\n')
curl http://localhost:8000/llm/v1/chat/completions \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "openai/gpt-4o",
    "messages": [{
      "role": "user",
      "content": [
        {"type": "text", "text": "What is in this image?"},
        {"type": "image_url", "image_url": {"url": "data:image/png;base64,BASE64_IMAGE_DATA", "detail": "auto"}}
      ]
    }],
    "max_tokens": 512
  }'

Replace BASE64_IMAGE_DATA with a PNG/JPEG base64 payload (no data: prefix). Works with vision-capable models such as openai/gpt-4o.