Loading page content
Configure a model call once and get the equivalent request as cURL, Python, JavaScript, or raw JSON, across five providers. Useful for the first call against a provider you have not used before.
Generate ready-to-use API requests for OpenAI, Anthropic, Google Gemini, DeepSeek, and Groq. Configure your parameters and copy the request in cURL, Python, JavaScript, or raw JSON.
curl -X POST 'https://api.openai.com/v1/chat/completions' \
-H 'Content-Type: application/json' \
-H 'Authorization: Bearer YOUR_API_KEY' \
-d '{
"model": "gpt-4.1",
"messages": [
{
"role": "system",
"content": "You are a helpful assistant."
},
{
"role": "user",
"content": "Explain Kubernetes pods in 3 sentences."
}
],
"temperature": 0.7,
"max_tokens": 1024
}'Every provider has a slightly different request shape. System prompts are a top-level parameter in some APIs and a message with role system in others. Stop sequences, max token limits, and streaming flags are all named differently. Getting the first call working is disproportionately annoying relative to how simple the request actually is.
No API key is required or accepted here: the tool generates the request text and you supply the credential yourself when you run it. Nothing is sent anywhere from this page.