Chat Completions

/v1/chat/completions accepts OpenAI Chat Completions style requests. Optional parameter support depends on the target model.

Basic request

curl https://ai.akaii.top/v1/chat/completions \
  -H "Authorization: Bearer $AKAI_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "MODEL_ID_FROM_CONSOLE",
    "messages": [
      { "role": "system", "content": "You are a concise assistant." },
      { "role": "user", "content": "Explain API gateways in one sentence." }
    ]
  }'

Streaming

When supported by the model, set:

{
  "stream": true
}

The client must consume Server-Sent Events as they arrive instead of waiting for one final JSON response.

Model support for temperature, tool calls, and multimodal input varies. Use the exact model ID returned by the dashboard or /v1/models.