Home › Inference Platforms › New API

New API - AI Inference Platforms Tool

Overview

New API is an open-source, next-generation LLM gateway and AI-asset management platform that aggregates multiple large-model providers behind a single, OpenAI-compatible interface. It exposes chat, responses, embeddings, image/audio/video, realtime and rerank endpoints while providing native format support for OpenAI Responses/Realtime, Anthropic Claude (Messages) and Google Gemini, enabling teams to call many providers with a single client integration. ([github.com](https://github.com/QuantumNous/new-api)) Beyond protocol unification, New API includes a web UI and admin console for token grouping, per-token/model quotas, usage analytics, online recharge and model charging (Stripe / local payment integrations), configurable billing strategies, and configurable "reasoning effort" (model-specific tuning and thinking-mode controls). Those management features make it suitable both for individual developers running a local gateway and for enterprises operating internal distribution or reseller channels. The project documents a Docker-first deployment, multi-instance (MySQL/Postgres + Redis) topology, and a set of environment variables for scaling and security. ([github.com](https://github.com/QuantumNous/new-api))

GitHub Statistics

  • Stars: 14,846
  • Forks: 2,918
  • Contributors: 192
  • License: NOASSERTION
  • Primary Language: JavaScript
  • Last Updated: 2026-01-07T12:52:27Z
  • Latest Release: v0.10.5

The repository is actively developed and widely adopted: the GitHub project lists roughly 14.8k stars and 2.9k forks, with thousands of commits and an active issues/PR queue, indicating significant community traction. Recent changelogs and frequent releases show an ongoing release cadence across late 2024–2025. ([github.com](https://github.com/QuantumNous/new-api)) Project governance looks collaborative: there are many contributors and a non-trivial backlog of issues and pull requests, and third-party repo analytics show steady issue and PR activity (open and closed) — evidence of an active maintainer/ contributor community. The project has also responded to security issues (an SSRF CVE was reported and patched in the 0.9.x series), which demonstrates responsiveness to vulnerabilities. ([issues.ecosyste.ms](https://issues.ecosyste.ms/hosts/GitHub/repositories/QuantumNous%2Fnew-api?utm_source=openai))

Installation

Install via docker:

git clone https://github.com/QuantumNous/new-api.git && cd new-api
docker-compose up -d
docker pull calciumion/new-api:latest
docker run --name new-api -d --restart always -p 3000:3000 -e TZ=Asia/Shanghai -v ./data:/data calciumion/new-api:latest
docker run --name new-api -d --restart always -p 3000:3000 -e SQL_DSN="root:123456@tcp(localhost:3306)/oneapi" -e TZ=Asia/Shanghai -v ./data:/data calciumion/new-api:latest

Key Features

  • Unified OpenAI-compatible gateway for OpenAI, Claude, Gemini, and custom channels.
  • Supports Responses, Realtime, Chat, Embeddings, Images, Audio, Video, and Rerank endpoints.
  • Token grouping, per-token quotas, and model-level permissioning for granular access control.
  • Online recharge and billing: Stripe and local payment integrations with flexible charging rules.
  • Intelligent channel routing with weighted selection, automatic failover, and retry logic.
  • Format conversion between OpenAI, Claude Messages, and Gemini chat formats.
  • Reasoning-effort controls (per-model thinking-mode and low/medium/high effort presets).
  • Visual dashboards, multi-language UI (EN/FR/JA/CN), and usage analytics for operators.

Community

New API has a large, active community and frequent releases; the GitHub repo shows high star/fork counts and an active issue/PR backlog, with many merged contributions and community-driven changelogs. The maintainers publish regular changelogs and release builds (desktop and container artifacts) and provide business/cooperation contacts for enterprise/custom deployments. The project has handled security reports (an SSRF issue was fixed in the 0.9.x line), but users should follow the docs for secure multi-instance deployment (SESSION_SECRET/CRYPTO_SECRET and Redis usage). For questions, bug reports, or enterprise inquiries the project links to GitHub Issues, Discussions, and business email contacts. ([github.com](https://github.com/QuantumNous/new-api))

Last Refreshed: 2026-01-09

Key Information

  • Category: Inference Platforms
  • Type: AI Inference Platforms Tool
Visit Official Website

Related Tools

Replicate

Platform to run and fine‑tune AI models with an API; host and deploy state-of-the-art models at scale.
HUGS

Optimized, zero‐configuration inference microservices from Hugging Face designed to simplify and accelerate the deployment of open AI models via an OpenAI‐compatible API.
OpenVINO Toolkit

Open-source toolkit to optimize and deploy AI inference across vision, speech, NLP, and generative models.
Dynamic Speculation

A novel method developed by Intel labs and Hugging Face that accelerates text generation by up to 2.7x using dynamic speculation lookahead in language models, integrated into the Transformers library.
LocalAI

Open‑source, OpenAI‑compatible local inference server for LLMs, TTS, ASR, and diffusion with CPU/GPU backends and container images.
Inference Endpoints by Hugging Face

A fully managed inference deployment service that allows users to easily deploy models (such as Transformers and Diffusers) from the Hugging Face Hub on secure, compliant, and scalable infrastructure. It offers pay-as-you-go pricing and supports a variety of tasks including text generation, speech recognition, image generation, and more.