Call Any Text Model Through One OpenAI-Compatible Endpoint
Basic conversation#
{
"model": "gpt-5.2",
"messages": [
{"role": "user", "content": "Explain vector databases in one sentence."}
]
}Add a system prompt#
{
"model": "claude-sonnet-4-5-20250929",
"messages": [
{"role": "system", "content": "You are a technical editor. Improve clarity of product copy."},
{"role": "user", "content": "Rewrite the button text 'Submission failed' to be friendly and actionable."}
]
}Run a multi-turn conversation#
{
"model": "gemini-3-flash-preview",
"messages": [
{"role": "user", "content": "Give me 3 customer support bot names."},
{"role": "assistant", "content": "1) HelpWave 2) CarePilot 3) SwiftSupport"},
{"role": "user", "content": "Add one style tag for each name."}
]
}Stream the output#
{
"model": "gpt-5.2",
"messages": [
{"role": "user", "content": "Generate 5 short titles (max 6 words) about AI video generation."}
],
"stream": true
}Know the request rules#
- Set
stream: truefor SSE streaming responses. - Different models support different context lengths, output limits, and optional parameters.
- For OpenAI-compatible workflows that need tools or multimodal structured input, use the Responses API.
See Error Codes for how failed requests are reported, and Authentication for the required Bearer token.