Create

Sign in to ReadmeX

or

How we handle your data: Privacy Policy

ReadmeX

Your AI reading companion

What would you like to catch up on?

Pick a briefing, or ask your own question.

Your community and people briefings stay personal to you.

new-api

new-api• Pinned

new-api: An AI gateway for models, applications, and agents.

What problem it solves

Teams using multiple upstream model services need a consistent client-facing API without reconfiguring every application whenever providers change. They also need centralized control over model routing, authorized access, usage, quotas, and costs.

Key features

  • Compatible model APIs — Expose OpenAI Chat Completions and Responses, Anthropic Messages, and Gemini APIs to clients, including streaming, tools, reasoning, and multimodal inputs where supported.

    Source: README

  • Configurable routing — Route requests across upstream services with model mappings, channel priorities and weights, retries, affinity, and multiple upstream keys.

    Source: README

  • Usage and cost management — Administer quotas, subscriptions, usage logs, cache accounting, and expression-based pricing for different usage tiers.

    Source: README

  • Access control and authentication — Manage users, groups, permissions, and API key restrictions, with OAuth/OIDC, passkeys, two-factor authentication, and login session management.

    Source: README

  • Task plugins and web console — Extend task APIs with JavaScript plugins and manage channels, models, usage, audit logs, and model tests through a multilingual web console.

    Source: README

Quick start

Copied verbatim from the project's README.

mkdir -p data
docker run --name new-api -d --restart unless-stopped \
  -p 127.0.0.1:3000:3000 \
  -e TZ=Asia/Shanghai \
  -v "$(pwd)/data:/data" \
  calciumion/new-api:latest
curl --fail-with-body http://localhost:3000/v1/responses \
  -H "Authorization: Bearer ${NEW_API_KEY}" \
  -H "Content-Type: application/json" \
  -d '{"model":"your-enabled-model","input":"Hello!"}'

Who it may suit

Inference: this may suit teams operating private multi-model services, developers building model-powered applications and agents, and organizations that need centralized access and cost management.

Topics to discuss

  • Which upstream providers and protocol combinations should receive priority in future compatibility work?

  • How can routing policies better balance availability, cost, and channel affinity across multiple upstream keys?

  • What plugin workflows would be most valuable for image, video, and other asynchronous task APIs?

Written by ReadmeX AI from the project's README, source and public GitHub data; it may contain mistakes.

Read the original README https://github.com/QuantumNous/new-api#readme

ReadmeX opened this community for a trending project; it is not run by the project. Maintainers can claim it from the community page.

github.comhttps://github.com/QuantumNous/new-api

Comments

…