NewApi — AI API Direct-Source Platform｜OpenAI/Claude/Gemini Relay, Open-Source Token Factory
NewApi is an open-source AI API gateway that connects 40+ upstream providers — OpenAI, Claude, Gemini, DeepSeek, Grok, and more — through a single OpenAI-compatible endpoint. Manage channels, API keys, quotas, and usage logs from one self-hosted control panel, while providing source Claude, GPT, Gemini and other API Tokens for relay platforms.
Supported providers:
OpenAI · Anthropic Claude · Google Gemini · DeepSeek · xAI Grok · Azure OpenAI · AWS Bedrock · Moonshot · Qwen · Zhipu GLM · Volcengine · Minimax · Midjourney · Cohere · Suno · Wenxin · Spark · Hunyuan · Xinference · and 20+ more
Provider Coverage for NewApi
NewApi connects to 30+ upstream AI providers: OpenAI · Anthropic Claude · Google Gemini · DeepSeek · xAI Grok · Azure OpenAI · AWS Bedrock · Moonshot · Qwen · Zhipu GLM · Volcengine · Minimax · Midjourney · Cohere · Suno · Wenxin · Spark · Hunyuan · Xinference · and more.
Why Teams Use NewApi
NewApi keeps model routing, channel control, API key management, quota rules, and usage visibility in one practical AI API gateway.
NewApi for routing and operations
Route multiple upstream AI models, review usage, and manage provider access through one NewApi deployment.
NewApi for OpenAI compatible access
Use OpenAI compatible API access while keeping channels, quotas, and usage inside one gateway workflow.
NewApi Core Features
Unified OpenAI-compatible endpoint
All upstream providers are exposed through a single NewApi endpoint that follows the OpenAI request/response format. Switch models or providers without changing application code.
Channel-based model routing
Define channels that map model names to specific upstream providers and keys. Configure load balancing weights, priority ordering, and automatic fallback rules across channels.
API key management and quota control
Issue user-facing API keys with per-key quota limits, rate limits, and group-based access policies. Upstream provider credentials are stored server-side in NewApi and never exposed to end users.
Usage logging and monitoring
Every request is logged with model, token count, latency, and status. Review per-user and per-channel consumption from the NewApi admin dashboard.
Self-hosted and open-source
NewApi is fully open-source. Deploy with Docker or as a standalone binary. Supports SQLite, MySQL, and PostgreSQL. Your data stays on your own infrastructure.
NewApi Use Cases
Teams use NewApi to power internal tools, public products, and multi-provider AI workflows without managing separate gateways for each provider.
AI application routing
Route chat, generation, and multimodal requests through one NewApi gateway with consistent access control.
Team platform operations
Operate shared AI infrastructure with model routing, provider access, API keys, and quota policies in one place.
OpenAI compatible migration
Keep OpenAI compatible API access while moving model switching and billing visibility into NewApi.
Multi-provider cost control
Track usage, review provider traffic, and keep quota policies visible across your AI API gateway operations.
What NewApi Can Do
NewApi gives teams one place to connect providers, standardize access, and keep AI gateway operations understandable.
Provider integration
Connect multiple AI providers and expose them through one consistent gateway entry point.
Quota and key control
Manage API keys, user groups, quotas, and access rules with a single operating model.
Usage visibility
Review request logs, usage records, and billing visibility without switching between provider consoles.
How NewApi Works
Connect providers
Add upstream providers and model routes to NewApi.
Control access
Manage API keys, quotas, groups, and channel rules in one control panel.
Monitor usage
Review logs, usage, and billing visibility across your AI API gateway.
Frequently Asked Questions about NewApi
What is NewApi?
NewApi is an open-source AI API gateway that aggregates 40+ AI providers behind a single OpenAI-compatible endpoint. NewApi handles model routing, API key management, quota control, and usage logging in one self-hosted deployment.
Which AI providers does NewApi support?
NewApi supports 40+ providers including OpenAI, Anthropic Claude, Google Gemini, DeepSeek, xAI Grok, Azure OpenAI, AWS Bedrock, Moonshot, Qwen, Zhipu GLM, Volcengine, Minimax, Midjourney, and more.
Is NewApi compatible with the OpenAI SDK?
Yes. NewApi exposes an OpenAI-compatible API. Change the base URL to your NewApi instance and use your NewApi key — no other code changes needed.
Can NewApi be self-hosted?
Yes. NewApi is fully open-source and built for self-hosted deployment. It supports SQLite, MySQL, and PostgreSQL, and can be deployed with Docker or as a standalone binary.
What is model routing in NewApi?
Model routing in NewApi lets you define channels — each channel maps a model name to a specific upstream provider and key. Configure load balancing, fallback rules, and priority weights so NewApi distributes requests automatically.
Start Using NewApi
Deploy NewApi, connect your providers, and manage AI API gateway operations from one place.
Open Console