XidaoApi

XiDao API Gateway

[](https://opensource.org/licenses/MIT) [](https://xidaoapi.com) [](https://xidaoapi.com) [](https://xidaoapi.com)

The affordable OpenAI-compatible AI API gateway for developers, startups, and AI agents.

Access 100+ LLM models through a single API endpoint — Claude 4.7, GPT-5.5, Gemini 3.0 Pro, DeepSeek V4, Qwen 3, Mistral Large, Llama 4, and more. Smart routing, automatic failover, and cost optimization built in.

Quick Start

from openai import OpenAI

client = OpenAI(
    api_key="your-xidao-api-key",
    base_url="https://api.xidaoapi.com/v1"
)

response = client.chat.completions.create(
    model="claude-4.7",
    messages=[{"role": "user", "content": "Hello!"}]
)
print(response.choices[0].message.content)

Open-Source Tools and Examples

| Repository | Description | |------------|-------------| | agent-eval-toolkit | Evaluate LLM agent outputs across providers — tool-use validation, regression testing, cost-aware scoring | | local-llm-router | Route prompts between local and cloud LLMs by task complexity — save 80%+ on AI costs | | xidao-python-examples | Python SDK examples — chat, streaming, embeddings, multi-provider switching | | xidao-nodejs-examples | Node.js/TypeScript — streaming, Vercel AI SDK, endpoint migration | | xidao-cookbook | Migration guides, failover patterns, cost control recipes | | llm-api-bench | Benchmark LLM providers on latency, throughput, cost, and quality | | llm-cost-calculator | Compare pricing across 50+ providers and estimate monthly costs | | llm-config-auditor | Audit OpenAI-compatible app configs for baseurl, model, retries, timeouts, and production-safety mistakes before deployment | | llm-failover-router-demo | Production failover router with circuit breakers and health checks | | mcp-cost-analyzer | Analyze MCP tool schema overhead, token usage, and per-turn cost for AI agents | | mcp-server-template | MCP server template with multi-provider routing and observability | | llm-provider-migration-checklist | Safe provider switching checklist with regression test suite | | llm-output-checker-action | GitHub Action to validate LLM responses in CI/CD pipelines | | llm-prompt-tester | Test and compare LLM prompts across providers — quality scoring, latency, cost |

from openai import OpenAI

client = OpenAI(
    api_key="your-xidao-api-key",
    base_url="https://api.xidaoapi.com/v1"
)

response = client.chat.completions.create(
    model="claude-4.7",
    messages=[{"role": "user", "content": "Privet!"}]
)
print(response.choices[0].message.content)