Skip to main content

Chat Completions

Use chat completions for message-based model calls.

POST /v1/chat/completions

Request​

curl https://api.voxvey.com/v1/chat/completions \
  -H "Authorization: Bearer $VOXVEY_TOKEN" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "deepseek/deepseek-v4-flash",
    "messages": [
      {
        "role": "system",
        "content": "Use only facts in the user message. Answer in one concise sentence."
      },
      {
        "role": "user",
        "content": "Facts: Voxvey API is a public provider gateway at https://api.voxvey.com. It exposes OpenAI-compatible model routes, Search routes, and MCP tools. Explain Voxvey API to a developer."
      }
    ],
    "stream": false
  }'

Example response​

{
  "id": "chatcmpl-...",
  "object": "chat.completion",
  "created": 1780711561,
  "model": "deepseek-v4-flash",
  "choices": [
    {
      "index": 0,
      "message": {
        "role": "assistant",
        "content": "Voxvey API is a public provider gateway at https://api.voxvey.com that exposes OpenAI-compatible model routes, Search routes, and MCP tools for developers."
      },
      "finish_reason": "stop"
    }
  ],
  "usage": {
    "prompt_tokens": 64,
    "completion_tokens": 120,
    "total_tokens": 184
  }
}

Required fields​

FieldTypeNotes
modelstringProvider-prefixed model ID
messagesarrayOpenAI-compatible chat messages

Notes​

  • Streaming is supported when the upstream model supports it.
  • Anthropic models use request and response translation behind the gateway.
  • BytePlus chat models use byte/* IDs and are forwarded to ModelArk.
  • Some models may include provider-specific metadata fields in the response.