gpt88 AI Relay: Unified API Access for OpenAI, Claude, and Gemini
gpt88 is a unified AI relay and large-model gateway for developers, covering OpenAI-compatible APIs, Claude-compatible routing, Gemini image generation, domestic routes, overseas routes, and transparent RMB-based billing.
What the gpt88 AI relay is
gpt88.cc acts as an AI relay and unified gateway for developers and teams. The goal is not to force a brand-new protocol. The goal is to normalize the most common access patterns into a more stable entry point.
Why teams use an AI relay
When a team uses GPT, Claude, Gemini, DeepSeek, Qwen, image models, and video models together, the vendor endpoints, auth styles, error formats, model naming, and network routes often differ. A relay reduces that integration sprawl.
- Use a consistent gateway for many model vendors.
- Choose domestic or overseas routing by environment.
- Reuse one setup pattern across SDKs, desktop tools, and workflow systems.
Core capabilities
- OpenAI-compatible routing at
https://api.gpt88.cc. - Claude / Anthropic-compatible routing through the standard API Base URL.
- Gemini image generation and video generation through the dedicated media route at
https://img.gpt88.cc. - Model navigation across chat, image, video, and audio categories.
- Integration docs for ChatBox, Cherry Studio, Claude Code, Codex CLI, Cursor, Cline, Dify, and more.
Base URL selection
The most common integration mistake is the wrong Base URL. OpenAI-style clients usually require/v1. Claude-style clients usually do not.
Standard API Base URL:
https://api.gpt88.cc
Direct image / video Base URL:
https://img.gpt88.cc
Standard APIs use /v1/...; native Gemini image APIs use /v1beta/...Minimal request example
Use cURL first to confirm that the relay path, key, and model are valid.
export GPT88_API_KEY="sk-your-gpt88-api-key"
curl https://api.gpt88.cc/v1/chat/completions \
-H "Authorization: Bearer $GPT88_API_KEY" \
-H "Content-Type: application/json" \
-d '{
"model": "gpt-5-2-chat-latest",
"messages": [
{"role": "user", "content": "Introduce the gpt88 AI relay in one sentence"}
]
}'For full SDK examples, continue with Python SDK, Node.js SDK, and cURL.
Which tools it fits
- OpenAI SDK, cURL, Cursor, and Cline through
https://api.gpt88.cc. - Claude Code and Anthropic-style SDKs through
https://api.gpt88.cc. - Gemini / NanoBanana image flows through the dedicated Gemini image API route.
- Workflow systems such as Dify through the OpenAI-compatible provider path.
Token power billing mindset
Traditional points-based relay:
top up → convert to virtual credits / points → apply multipliers → manually reverse real cost
gpt88.cc:
1 RMB topped up = 1 RMB balance
actual deduction (RMB) = official usage (USD) x selected group multiplier
group multipliers are shown in the API Keys group selector
lower multipliers mean lower unit costNext steps
- Quickstart for the first real request.
- Complete Integration Guide for end-to-end setup.
- Integration Guides for tool-specific setup.
- Auth & Billing for balance and usage behavior.