Guides
Tutorials and comparisons for an AI gateway: call Claude, GPT, Gemini, DeepSeek and more through one OpenAI-compatible endpoint, how to choose a gateway or relay service, and how to set up Claude Code and Codex.
How to Choose an AI API Relay Service: An 8-Point Checklist and How to Verify One Yourself
What the Chinese-community term "AI relay" (zhongzhuan, 中转站) means, how a relay or reseller differs from an official API and from an AI gateway, and an 8-point checklist covering billing, model authenticity, data policy, stability, invoices, compliance, compatibility and exit cost.
What Is an AI Gateway? When You Need One and How to Choose
A plain explanation of AI gateways (LLM gateways): what they do, the problems they solve, when you can skip one, and a checklist for choosing between self-hosted and hosted options.
Claude Code Alternative: Official Subscription vs. a Pay-As-You-Go API Gateway
An honest comparison of using Claude Code on Anthropic's subscription versus pointing it at an API gateway billed per token, with the trade-offs and who each option fits.
How to Use Claude Code with a Custom API Endpoint (OriginCoder Setup Guide)
Step by step: point Claude Code at OriginCoder with environment variables, settings.json or one-click CC-Switch import, then verify the connection and fix the common errors.
Claude Code Proxy: HTTP Proxy or API Gateway? Which One You Need
The term Claude Code proxy can mean two different things: routing traffic through an HTTP proxy, or sending requests to a different API address. How to tell them apart and set up each one.
Claude Code Usage Limits Explained: Why You Hit Them and What to Do
What Claude Code subscription usage limits are, how they usually show up, and your options when you run into them, including pay-as-you-go API access, with honest trade-offs.
How to Use Codex CLI with a Custom API Endpoint (OriginCoder Setup Guide)
Configure Codex CLI to use OriginCoder's OpenAI-compatible endpoint through config.toml and an environment variable, or by one-click CC-Switch import, then verify it and fix common errors.
DeepSeek API Through an OpenAI-Compatible Gateway: Python, Node and curl
Call DeepSeek models through an OpenAI-compatible gateway with one key. Working Python, Node.js and curl examples, how to find the model name, and the common errors.
Gemini API Through an OpenAI-Compatible Gateway: curl, Python and Node
Call Gemini models through an OpenAI-compatible gateway with one key. Working curl, Python and Node.js examples, the Gemini-native paths, how to find the model name, and the common errors.
LLM Gateway Comparison: LiteLLM, Portkey, OpenRouter and OriginCoder
How a self-hosted gateway (LiteLLM), hosted gateways (Portkey, OpenRouter) and OriginCoder differ, and how to choose an AI gateway for your team without the marketing.
What Is an OpenAI-Compatible API? Switch Providers with Two Settings, and Debug the Differences
What an OpenAI-compatible API is, how changing only the Base URL and key switches model providers, common differences in chat completions, streaming and tool calls, and how to debug 401, 404, model names and buffered streams.
OpenRouter Alternatives: Hosted Aggregators, LiteLLM and OriginCoder Compared
How to choose an OpenRouter alternative: hosted aggregator APIs, a self-hosted LiteLLM proxy, or a gateway such as OriginCoder. What each is, who it fits, and what to check, with a disclosure.
OriginCoder Quick Start: Create a Key and Connect Claude Code with CC Switch
From creating an API key to one-click import in CC Switch, enabling the profile and using OriginCoder in Claude Code, with the Claude and GPT Base URLs explained and a screenshot for every step.