mingyan

OpenAI Completions API

OpenAI Chat Completions-compatible text generation API. Currently only deepseek-v4-flash is available.

POST/gw_llm_power/v1/chat/completions

baseUrlhttp://apillm.globalaiopc.com/gw_llm_power

This endpoint is compatible with the OpenAI Chat Completions request and response format, so it can be used with OpenAI SDKs or compatible clients.

OpenAI Chat Completions reference: https://platform.openai.com/docs/api-reference/chat/create

Authentication

Get Key

Use the Authorization header for authentication.

Required headers:

cURL

Authorization: Bearer {{YOUR_API_KEY}}
Content-Type: application/json

Supported Models

Models are grouped by series below:

DeepSeek

Model	Image Analysis	Video Analysis	Notes
`deepseek-v4-flash`	Not supported	Not supported	Suitable for text chat, generation, summarization, rewriting, and coding assistance

Zhipu GLM

No available models yet. Models will be added here after launch.

Doubao

No available models yet. Models will be added here after launch.

Request Parameters

modelstringrequired

Model name to use. Currently set this to deepseek-v4-flash.

messagesarrayrequired

Conversation message array. Each message object usually contains role and content.

temperaturenumber

Sampling temperature to control output randomness. Recommended range is 0 to 2.

top_pnumber

Nucleus sampling parameter. It is recommended to tune either temperature or top_p, not both.

streamboolean

Whether to enable streaming responses. When set to true, the response is returned as SSE and ends with data: [DONE].

max_completion_tokensinteger

Maximum output token count.

stopstring/array

Stop sequence(s). Can be a string or an array.

Messages Array Structure

rolestringrequired

Message role. Supported values: system, user, assistant.

contentstringrequired

Message content. deepseek-v4-flash currently supports plain text only and does not support multimodal arrays for image or video analysis.

Response

Returns an OpenAI-compatible response format.

idstring

Unique identifier for the chat request.

objectstring

Object type, usually chat.completion.

createdinteger

Response creation timestamp.

modelstring

Actual model name used.

choices[].message.rolestring

Model response role, usually assistant.

choices[].message.contentstring

Generated text content.

usage.prompt_tokensinteger

Input token count.

usage.completion_tokensinteger

Output token count.

usage.total_tokensinteger

Total token count.

Notes

The request URL is http://apillm.globalaiopc.com/gw_llm_power/v1/chat/completions
Pass the API key using the Authorization header
deepseek-v4-flash does not support image analysis or video analysis. Send plain text messages only
Streaming responses use SSE and end with data: [DONE]