XiuRouter is a model API service for developers connecting applications and AI agents to leading models through one API. It supports OpenAI Responses and Chat Completions, Anthropic Messages, and Gemini GenerateContent. Scoped API keys, request-level usage and cost records, and guided setup for 14 apps help teams integrate and operate model access with clearer controls.
Developers, AI engineering teams, startups, and businesses integrating multiple model providers into applications, coding agents, internal tools, and automated workflows.
Features & Use Cases
- One API for leading model providers
- OpenAI Responses and Chat Completions support
- Anthropic Messages support
- Gemini GenerateContent support
- Scoped API keys with model, budget, and expiration controls
- Request-level usage and cost records
- Developers, AI engineering teams, startups, and businesses integrating multiple model providers into applications, coding agents, internal tools, and automated workflows.
Pros & Cons
- Native support for four major API protocol families
- Centralized keys, budgets, usage, and cost visibility
- Guided Agent integration workflow for 14 apps
- Selected models and service groups can save over 90% versus provider reference prices
- Paid only, with no free plan or free trial
- Pricing and availability vary by model and service group
- Built for API and Agent workflows rather than a no-code consumer chat interface
Full Review
Overview
XiuRouter is a usage-billed model API service for developers connecting applications, coding agents, and internal tools to Claude, GPT, Gemini, Grok, and other models through one account and API-key system. XiuAI and its products are operated by XiuLab Inc, a U.S. corporation.
API compatibility
The service documents four protocol families: OpenAI Responses, OpenAI Chat Completions, Anthropic Messages, and Gemini GenerateContent. OpenAI-style clients use https://router-api.xiu.ai/v1, while Messages and Gemini clients use https://router-api.xiu.ai. The client’s request format determines the protocol, so the Base URL, request path, exact model ID, and service tier must match the selected client or SDK.
API keys and usage records
Each API key can be configured with a service tier, model scope, credit limit, and expiration. After a request, the Usage view records the model, tier, input and output tokens, cache usage, status, latency, and settled cost. This gives developers request-level information for reconciliation, monitoring, and spend controls.
Guided integrations
XiuRouter provides protocol-aware setup for 14 clients and frameworks: ChatGPT and Codex, Claude Code, Cursor, DeepSeek Harness, OpenCode, OpenWork, Cline, Continue, OpenClaw, Hermes Agent, Aider, Open WebUI, LibreChat, and Vercel AI SDK. The documentation recommends completing a real request and checking its Usage record after configuration.
Pricing
XiuRouter is a paid-only service with usage-based billing; the listing does not advertise a free plan or free trial. Rates vary by model, service tier, and actual input, output, cache-read, and cache-write usage, while models billed per request are labeled separately. The live catalog compares each applicable XiuRouter rate with a provider reference price. Selected models and service groups currently show savings above 90%, but that discount is not uniform and can change with the public catalog.
Compatibility boundaries
XiuRouter provides model API access rather than every feature of a provider’s hosted platform. Its Responses endpoint is stateless, so stored conversations, background mode, provider-hosted web search, and provider-hosted file search are not included. Anthropic-hosted server tools, Message Batches, and Files remain Anthropic platform features. Tool calling, streaming, structured output, and client-specific agent behavior should be verified with the selected model and target workflow.
Who it is for
The documented use cases cover coding agents, existing OpenAI-compatible applications, multi-model workloads, and internal systems that need scoped access and per-request cost records. Before production use, XiuRouter recommends confirming the protocol, Base URL, model ID, tier, and current rate, then sending a small request and reviewing its recorded result.
Visit XiuRouter
View current pricing and documentation on the official site.
