Chat with a language model
OpenAI's Chat Completions, for every chat model in GET /models (Claude, GPT, Gemini, Grok, DeepSeek, Qwen, Llama, Mistral and more): point any OpenAI SDK at https://api.cut.pro/api/v1 with your API key. Streaming, tools, images in the messages and structured answers work where the model supports them. Billed by the tokens read and written at the prices of GET /models; fractions of a credit add up and leave the balance as whole credits. A plain answer reports what it cost in usage.credits. The balance must cover what the call can cost, counting the answer's max_tokens.
Authorization
apiKey In: header
A Chat Completions request as OpenAI defines it; every field goes to the model as written.
TypeScript Definitions
Use the request body type in TypeScript.
A Chat Completions request as OpenAI defines it; every field goes to the model as written.
Response Body
application/json
application/json
application/json
application/json
curl -X POST "https://example.com/chat/completions" \ -H "Content-Type: application/json" \ -d '{ "model": "string", "messages": [ { "property1": null, "property2": null } ], "property1": null, "property2": null }'{ "code": "REQUEST_INVALID"}Get the download link of a generation GET
Signed link that saves the full-quality file, valid for 6 hours. Only for a `ready` generation. Once downloaded, its credits can no longer be asked back.
Turn text into embeddings POST
OpenAI's embeddings request, for every model with `endpoint: embeddings` in `GET /models`. Billed by the tokens read; `usage.credits` reports what it cost.