Your first request

A complete non-streaming request looks like this.

curl https://api.proxy.gonka.gg/v1/chat/completions \
  -H "Content-Type: application/json" \
  -H "Authorization: Bearer sk-your-api-key-here" \
  -d '{
    "model": "MiniMaxAI/MiniMax-M2.7",
    "messages": [
      {"role": "system", "content": "Be brief."},
      {"role": "user", "content": "What is Gonka in one sentence?"}
    ]
  }'

The response is OpenAI-shaped: id, choices[0].message.content, usage.prompt_tokens, usage.completion_tokens.

Model IDs

Copy the id from GET /v1/models. Current families:

• deepseek-ai/DeepSeek-V4-Flash-0731
• MiniMaxAI/MiniMax-M2.7
• zai-org/GLM-5.3-Flash

These chat models run on Gonka. Jev (jev) is not a chat model and is not a Gonka model. Short names like gpt-4o are not aliases here.

What gets billed

Prompt + completion tokens, same USD rate. See Pricing. Empty or failed requests that never reach a host are usually not billed; 401/402/404 on the proxy side do not spend tokens.

Streaming variant

Add "stream": true and read SSE data: lines. Details on Streaming.

If it fails