Using API and Playground

Use the playground to test and experience AI models online, and call New API via OpenAI-compatible interfaces.

Playground

The playground is used for simple online testing and experiencing AI models. Users can input questions or instructions, interact with the model, and view the generated results.

Select Group and Model: In the bottom right of the input box, select the available model group and specific model to confirm the model used for this conversation.
Set Request Parameters: Click the parameter settings button to enable and adjust generation parameters such as Temperature, Top P, Frequency Penalty, Presence Penalty, and Max Tokens as needed.
Fill in Conversation Content: Enter questions, instructions, or prompts in the bottom input box; search, upload files, upload photos, screenshot, and photo-taking functions are not yet implemented.
Send Request: After confirming the model, parameters, and input content, click the "Send" button to submit the request.
View Response Results: After the model finishes generating, you can view the returned content in the conversation area and continue to input new questions for multi-turn conversations.

Playground Page

API Address and Authentication

The OpenAI-compatible address for the gateway is typically:

https://your-host/v1

Use your API key as a Bearer Token:

Authorization: Bearer sk-your-api-key

For example, with chat completions:

curl https://your-host/v1/chat/completions \
  -H "Content-Type: application/json" \
  -H "Authorization: Bearer sk-your-api-key" \
  -d '{"model":"gpt-4o-mini","messages":[{"role":"user","content":"你好"}]}'

SDK Examples

Python (OpenAI SDK)

from openai import OpenAI

client = OpenAI(
    api_key="sk-xxxxxxxx",
    base_url="https://your-host/v1",
)

response = client.chat.completions.create(
    model="gpt-4o-mini",
    messages=[{"role": "user", "content": "你好"}],
)
print(response.choices[0].message.content)

Claude Native Format

curl https://your-platform.com/v1/messages \
  -H "x-api-key: sk-xxxxxxxx" \
  -H "anthropic-version: 2023-06-01" \
  -H "content-type: application/json" \
  -d '{"model": "claude-3-5-sonnet-20241022", "max_tokens": 1024, "messages": [{"role": "user", "content": "Hello"}]}'

Gemini Native Format

curl "https://your-platform.com/v1beta/models/gemini-1.5-pro:generateContent?key=sk-xxxxxxxx" \
  -H "Content-Type: application/json" \
  -d '{"contents": [{"parts": [{"text": "Hello"}]}]}'

Supported API Endpoints

Interface	Path	Description
Chat Completions	`POST /v1/chat/completions`	OpenAI Chat Completions compatible interface, requires `model` and `messages`, supports streaming or regular responses.
Playground Chat	`POST /pg/chat/completions`	Web playground specific interface, uses login session to select group and model, and reuses unified traffic splitting, protocol conversion, billing, and retry mechanisms.
Text Completions	`POST /v1/completions`	Traditional text completions interface, requires `prompt` and `model`.
Vector Embeddings	`POST /v1/embeddings`	Converts text or arrays of text into vectors, requires `input`, converted to upstream format by channel adapter.
Gemini Embeddings Compatible Interface	`POST /v1/engines/{model}/embeddings`	Compatible with older Gemini-style embedding paths, processes embedding requests based on the model name in the URL.
Image Generation	`POST /v1/images/generations`	Generates images based on prompts; specific model, size, quality, and quantity depend on request parameters and configured channels.
Image Editing	`POST /v1/images/edits`	Edits images based on input image and prompt, supports JSON and `multipart/form-data` requests.
Legacy Edit Interface	`POST /v1/edits`	Reserved compatible path, code enters image Relay processing; new clients are advised to prioritize using `/v1/images/edits`.
Speech to Text	`POST /v1/audio/transcriptions`	Converts audio to text, forwards to corresponding speech recognition channel according to OpenAI audio interface format.
Audio Translation	`POST /v1/audio/translations`	Recognizes audio content and outputs translated text.
Text to Speech	`POST /v1/audio/speech`	Converts input text to audio.
Rerank	`POST /v1/rerank`	Reranks candidate `documents` based on `query` for relevance, requires non-empty query content and document list.
Content Moderation	`POST /v1/moderations`	Performs content moderation on text or other inputs; if no model is provided, a default moderation model candidate will be used.
Responses API	`POST /v1/responses`	OpenAI Responses format interface, supports capabilities such as input, tools, and response events.
Responses Compression	`POST /v1/responses/compact`	Performs compression processing on Responses request context.
Codex Web Search	`POST /v1/alpha/search`	Codex independent web search format interface, currently only allows processing by Sub2API, New API, Codex, and advanced custom channels that support this capability.
Claude Messages	`POST /v1/messages`	Anthropic Claude Messages compatible interface, requires `model` and message content, converted to upstream protocol by New API when necessary.
Realtime Conversation	`GET /v1/realtime` (WebSocket)	OpenAI Realtime style bidirectional real-time communication interface, models are usually passed via query parameters, and API keys can be passed via Bearer Token or WebSocket subprotocol.
Gemini Native Model Invocation	`POST /v1beta/models/{model}:{action}`	Gemini native format interface, supports model operation paths, such as `generateContent`, `streamGenerateContent`, and embedding-related operations.
Gemini Compatible Model Invocation	`POST /v1/models/{model}:{action}`	Accepts Gemini-style model operation requests with the `/v1` prefix, parsed and converted by Gemini Relay.
Model List	`GET /v1/models`	Returns OpenAI format model list by default; returns Claude format when using Anthropic request headers, and Gemini format when using `x-goog-api-key` or `key` query parameters.
Model Details	`GET /v1/models/{model}`	Queries single model information; returns `model_not_found` error object if the model does not exist.
Gemini Model List	`GET /v1beta/models`	Returns a list of available models in Gemini format, response fields include `models` and `nextPageToken`.
OpenAI Format Model List	`GET /v1beta/openai/models`	Returns a model list in OpenAI format under the Gemini compatible namespace.