Text Generation Overview

The text generation endpoint is fully compatible with the OpenAI Chat Completions API. The platform has integrated multiple mainstream large models. Each model family may have two calling styles:

1. OpenAI-compatible format (recommended): uniformly use /v1/chat/completions, only change the model name
2. Native format (optional): some models also support their official native API format, such as Claude's /v1/messages or Gemini's /v1/models/{model}:generateContent

We recommend using the OpenAI-compatible format first, so switching models later requires no code changes.

Integrated Model Families

Model Family	Description	Recommended Endpoint
OpenAI (GPT)	Full GPT-5.x series, o-series reasoning models	POST /v1/chat/completions
Claude	Claude Opus / Sonnet / Haiku series, native format supported	POST /v1/chat/completions or Anthropic native
Gemini	Google Gemini 2.5 / 3.x series, native format supported	POST /v1/chat/completions or Gemini native
DeepSeek	DeepSeek V4 Pro / Flash	POST /v1/chat/completions
Kimi (Moonshot)	Moonshot Kimi K3	POST /v1/chat/completions
Zhipu AI (GLM)	Zhipu GLM series	POST /v1/chat/completions
Grok (xAI)	Grok 4.5 / 4.6, 500k context	POST /v1/chat/completions

Generic Example

curl $URL/v1/chat/completions \
  -H "Authorization: Bearer $YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "gpt-4o",
    "messages": [
      {"role": "system", "content": "You are a helpful assistant"},
      {"role": "user", "content": "Introduce yourself"}
    ],
    "stream": false
  }'

Common Parameters

Parameter	Type	Required	Description
model	string	Yes	Model name; see each family page for details
messages	array	Yes	Conversation message list
stream	boolean	No	Whether to stream the response, default false
temperature	number	No	Sampling temperature 0–2
max_tokens	integer	No	Maximum tokens to generate
top_p	number	No	Nucleus sampling

Full parameters are consistent with the official OpenAI documentation.