LLM API • LOAD TEST • CLIENT-SIDE
LLM Concurrency
Test
Fires N simultaneous requests and measures whether they run in parallel. Local mode automatically relays API calls through this machine to avoid browser CORS failures.
Your API key stays in this browser. On the local deployment, requests go through the local relay automatically, so providers do not need browser CORS support. Fill in the API key and endpoint, then leave the field to load available models automatically.
Test Configuration
Base URL is auto-resolved to the full endpoint on blur (OpenAI → /chat/completions, Anthropic → /v1/messages). A full URL you already pasted is left untouched.
Mapped to each provider's native reasoning or thinking setting.