LLM API Tester
Configure an endpoint, API key and model — pick from common providers like OpenAI, DeepSeek, Qwen and Ollama, or paste a custom URL — then send a test prompt to verify the model responds. The request goes straight from your browser to the model service. On the side, the tool generates the equivalent request code in curl, Python requests, Node.js fetch, Go and PowerShell, so you can run the exact same call from your own terminal or application.
Frequently asked questions
Is my API key uploaded to this site?
No. The request is sent directly from your browser to the model provider; your key never passes through this site and only stays in the current page.
Why does the browser sometimes block the request?
Some providers do not allow cross-origin (CORS) calls from a browser. If that happens, use the generated curl/Python/Node/Go/PowerShell code in your own terminal instead.
Related tools
Token Counter
Count tokens with real OpenAI tokenizers (cl100k_base / o200k_base).
LLM API Cost Calculator
Estimate any LLM API cost from token counts and your own per-1M-token prices.
Chat to OpenAI Messages
Convert a plain chat transcript into OpenAI messages JSON.
JSON Schema Generator
Generate JSON Schema draft-07 from a JSON sample for function calling.