Find the right AI for you

Search by name or tell us what you want to do
API servicesChoose how to access modelsAI modelsCapabilities, specifications, and pricingAI toolsFind useful tools by taskMCP directoryConnect tools, data, and workflowsAgent skillsFind reusable skills for AI agentsBenchmarksCompare model performanceOnline toolsConvert accounts, format JSON, estimate costs and test APIsModel timelineExplore model releases (中文)Use casesStart with a practical task (中文)
EnterSearch · ↑↓Choose a resultEscClose
Submit中文
API CONNECTION LAB

API connectivity tester

Test OpenAI Chat, Responses, Anthropic Messages and Gemini protocols directly from your browser. Sends no requests by default; clicking sends one test that may incur charges.

API CONNECTION LAB

API connectivity tester

Choose a protocol, check the destination and send one text request.

Request destination · POSTEnter a base URL and model to see the full destinationYour Key and prompt go only to this destination. Each click sends at most one generation request without retries; the browser may also send a CORS preflight.

Tests may incur charges. A token limit is not a spending cap. Timing out or stopping reception may still incur costs. A low limit may stop a reasoning model before it outputs text.

The destination must allow browser cross-origin requests. We do not proxy requests. A missing browser response does not prove service unavailability.

Checks one text response for the selected protocol, without verifying model identity, balance or long-term reliability. OpenAI Chat uses max_completion_tokens, Responses uses max_output_tokens, Anthropic uses max_tokens and Gemini uses maxOutputtokens. Unsupported parameters retain the actual error; no automatic protocol fallback or retry. Reception is limited to 128 KiB. Changing or leaving the page does not revoke a request already received by the server.

View model monitoring → Browse API services →