Developer Center

One place for every API this site offers.

This site provides OpenAI-compatible, Anthropic (Claude)-compatible, Google and OpenRouter relay APIs, plus site-native utility APIs (Google Search / News, chat, credits) and an MCP server. Usage is billed to your account credits; failed calls are never charged.

Get your API key API Pricing

Quick Start

curl https://www.myai168.com/en/api/openai/v1/chat/completions \
  -H "Authorization: Bearer <API_KEY>" \
  -H "Content-Type: application/json" \
  -d '{"model":"gpt-5.6-sol","messages":[{"role":"user","content":"Hello"}]}'

List the models each relay accepts with GET https://www.myai168.com/en/api/openai/v1/models or GET https://www.myai168.com/en/api/anthropic/v1/models — the returned list is exactly what the relay accepts; other models are rejected with No pricing info! Please use other models!.

Base URLs

OpenAI-compatible     https://www.myai168.com/en/api/openai/v1
Anthropic (Claude)    https://www.myai168.com/en/api/anthropic/v1
Google                https://www.myai168.com/en/api/google/v1beta
OpenRouter-compatible https://www.myai168.com/en/api/openrouter/v1

Authentication

Create an API key on the Developer API Key page, then send it with every request:

Authorization: Bearer <API_KEY> — works on all endpoints (recommended).
x-api-key: <API_KEY> — also accepted on Anthropic (Claude) endpoints (SDK compatible).
x-goog-api-key: <API_KEY> — also accepted on Google endpoints (SDK compatible).
X-Dev-Key: <API_KEY> — also accepted on the site-native Google Search / News, chat and MCP endpoints.

Never place API keys in URL query strings — URLs are recorded in access logs and browser history. Endpoints do not accept keys via query parameters.

Endpoints

OpenAI-compatible relay

Endpoint	Method	Description
`https://www.myai168.com/en/api/openai/v1/models`	GET	List supported OpenAI models (generated from the pricing table).
`https://www.myai168.com/en/api/openai/v1/models/{model}`	GET	Retrieve one model (SDK `client.models.retrieve` compatible).
`https://www.myai168.com/en/api/openai/v1/chat/completions`	POST	Chat Completions relay (streaming supported; `web_search_options` billed per use).
`https://www.myai168.com/en/api/openai/v1/responses`	POST	Responses API relay (streaming supported; `web_search` tool billed per use).
`https://www.myai168.com/en/api/openai/v1/embeddings`	POST	Embeddings relay (OpenAI-compatible request/response; accepted models: `voyage-4`, `voyage-4-large` — billed per input token).
`https://www.myai168.com/en/api/openai/v1/images/generations`	POST	Image generation relay.
`https://www.myai168.com/en/api/openai/v1/audio/speech`	POST	Text-to-speech relay (returns audio binary).
`https://www.myai168.com/en/api/openai/v1/audio/transcriptions`	POST	Speech-to-text relay (multipart upload).
`https://www.myai168.com/en/api/openai/v1/audio/translations`	POST	Speech translation to English (multipart upload).
`https://www.myai168.com/en/api/openai/v1/audio/voices`	GET	List available built-in TTS voices (free).
`https://www.myai168.com/en/api/openai/v1/audio/voices`	POST	Custom voice creation relay.

Anthropic (Claude) relay

Endpoint	Method	Description
`https://www.myai168.com/en/api/anthropic/v1/models`	GET	List supported Anthropic (Claude) models.
`https://www.myai168.com/en/api/anthropic/v1/models/{model_id}`	GET	Retrieve one model (SDK `models.retrieve` compatible).
`https://www.myai168.com/en/api/anthropic/v1/messages`	POST	Messages relay (streaming, prompt caching and thinking supported; `web_search` server tool billed per use; other server-side / hosted tools such as code execution are not supported — see the guide).
`https://www.myai168.com/en/api/anthropic/v1/messages/count_tokens`	POST	Token counting relay (free).

Google relay

Endpoint	Method	Description
`https://www.myai168.com/en/api/google/v1beta/models`	GET	List supported Google models (generated from the pricing table).
`https://www.myai168.com/en/api/google/v1beta/models/{model}`	GET	Retrieve one model.
`https://www.myai168.com/en/api/google/v1beta/models/{model}:generateContent`	POST	Gemini chat / vision / image generation relay.
`https://www.myai168.com/en/api/google/v1beta/models/{model}:streamGenerateContent`	POST	Gemini streaming relay (SSE).
`https://www.myai168.com/en/api/google/v1beta/models/veo-*:predictLongRunning`	POST	Veo video generation (long-running operation).
`https://www.myai168.com/en/api/google/v1beta/models/lyria-*:predict`	POST	Lyria music generation.
`https://www.myai168.com/en/api/google/v1beta/operations/{name}`	GET	Poll a long-running operation (free).

OpenRouter relay

Endpoint	Method	Description
`https://www.myai168.com/en/api/openrouter/v1/models`	GET	List OpenRouter models (proxied from the official OpenRouter list, cached 10 minutes).
`https://www.myai168.com/en/api/openrouter/v1/chat/completions`	POST	Chat Completions relay to OpenRouter (hundreds of `vendor/model` ids).
`https://www.myai168.com/en/api/openrouter/v1/chat/responses`	POST	Same Chat Completions request/response format as above (NOT the OpenAI Responses API schema); behaves the same as chat/completions.
`https://www.myai168.com/en/api/openrouter/v1/audio/speech`	POST	Text-to-speech relay.
`https://www.myai168.com/en/api/openrouter/v1/audio/transcriptions`	POST	Speech-to-text relay (multipart upload).
`https://www.myai168.com/en/api/openrouter/v1/video/generations`	POST	Video generation relay.

Site-native APIs

Endpoint	Method	Description
`https://www.myai168.com/en/api/google/search`	GET	Google web search (structured JSON results).
`https://www.myai168.com/en/api/google/news`	GET	Google News search (JSON results with resolved article URLs).
`https://www.myai168.com/en/api/user/get_credits`	GET	Query your remaining credits (free).
`https://www.myai168.com/en/ai/chat/chat`	POST	Site-native chat endpoint (SSE streaming; any chat model enabled on this site).

Not supported

The following vendor endpoints are NOT provided by this relay — requests to them return 404 (or 400 for hosted tools inside a supported endpoint). Do not guess from the SDK surface:

OpenAI /v1/batches (Batch API discount tier), /v1/files, /v1/fine_tuning, /v1/assistants, /v1/vector_stores.
OpenAI hosted tools inside Responses (code_interpreter / file_search / image_generation / computer use) — rejected with 400; only web_search is supported.
Anthropic (Claude) /v1/files, /v1/messages/batches, and server-side tools other than web_search (code execution is rejected with 400) — the relay uses one shared upstream workspace, so file / container APIs would risk cross-user access.

Streaming

Chat Completions, Responses, Anthropic Messages and Gemini streamGenerateContent support streaming. Send "stream": true (or use the streaming endpoint) and read the response as Server-Sent Events (data: {JSON} lines). With curl, add -N (--no-buffer) to see tokens as they arrive.

Billing and response headers

Usage is billed to your account credits. Non-streaming relay responses include X-Cost-Points (points charged for this call) and X-Credits-Remaining (your balance after the call). Streaming responses cannot carry these headers; instead, after the final event (e.g. data: [DONE]) the stream ends with SSE comment lines carrying the same information:

: cost=12
: credits-remaining=34567

Comment lines (starting with :) are ignored by spec-compliant SSE parsers and official SDKs, so they never affect compatibility; read them from the raw stream if you want per-call billing. You can always query GET https://www.myai168.com/en/api/user/get_credits. Failed calls, and calls where the vendor returns no usage information, are never charged.

Error handling

HTTP status	Meaning
`401`	Missing or invalid API key. Repeated failures from the same IP are rate-limited.
`402`	Insufficient credits (`Insufficient credits. Please top up!` / `You need to top up!`).
`400` + `No pricing info! Please use other models!`	The requested model is not on this relay's supported list — call the `/models` endpoint for the accepted list.
`400`	Invalid request body, or an unsupported hosted / server-side tool was requested.
`502`	Upstream vendor error — never charged.

Documentation

API Documentation

OpenAI ChatGPT API — Chat Completions / Responses / audio / images relay, request parameters, errors, streaming.
Anthropic (Claude) API — Messages relay, request parameters, errors, streaming.
Google Gemini API — generateContent / streamGenerateContent / Veo / Lyria relay.
Google Search API — Google web search (structured JSON results).
Google News API — Google News search (JSON results with resolved article URLs).
OpenRouter API — Multi-vendor LLM gateway relay (Chat Completions format).
API Pricing — How usage is converted to account credits.

CLI & Agent Guides

MYAI168 API & MCP

MYAI168 MCP CLI — Connect CLI / coding agents to this site via MCP with an API key.
MYAI168 MCP Web & App — Connect Claude.ai and other MCP clients via OAuth.

Tools

Your IP and Country — Shows the IP address you are connecting from and the country it is located in.

MCP server

Chat modules this site shares over MCP are exposed as MCP (Model Context Protocol) tools. Connect an MCP client to:

https://www.myai168.com/en/mcp/

Authentication: Authorization: Bearer <API_KEY> (CLI / coding agents), or OAuth 2.1 login for clients such as Claude.ai (leave Client ID / Secret empty — the server supports Dynamic Client Registration). See the MCP guides in the documentation list above.