# yym68686/uni-api

This is a project that unifies the management of LLM APIs. It can call multiple backend services through a unified API interface, convert them to the OpenAI format uniformly, and support load balancing. Currently supported backend services include: OpenAI, Anthropic, DeepBricks, OpenRouter, Gemini, Vertex, etc.

Repository: https://github.com/yym68686/uni-api
Canonical: https://ross.abutalabs.com/products/uni-api
Homepage: https://0-0.pro/r/uniapi
Language: Python
License: Apache-2.0
License Family: permissive
Topics: claude, openai, uni-api, gemini, vertex, aws, grok, azure
Last push: 2026-08-26T18:12:06+00:00

## Health v2 (maintenance only)
Score: 86/100 (v2, computed 2026-09-03T02:39:23.370411+00:00)
- activity 99, release rhythm 87, longevity 56
- inputs: {"age_days": 790, "days_push": 7, "days_rel": 7, "gap_med": 0, "n_releases_24m": 242}
- flags: none
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 1257, forks 156 (observed 2026-08-28T04:04:09.357276+00:00)

## What it is
uni-api is a self-hosted Python service that unifies management of multiple LLM provider APIs behind a single OpenAI-compatible endpoint. It converts requests to/from providers like OpenAI, Anthropic, Gemini, Vertex, Azure, AWS, and OpenRouter, with load balancing, automatic retry, channel cooling, and fine-grained API key permissions configured via a YAML file.

## Use cases
- unify multiple llm api keys behind one openai-compatible endpoint
- load balance requests across several llm providers
- convert anthropic and gemini apis to openai format
- self-host a personal llm api gateway without a frontend
- rotate and retry across llm api channels automatically
- control which models each api key can access
- use one base url for claude, gpt, and gemini in developer tools

## When to choose
- you want a lightweight, config-file-only LLM gateway for personal use
- you need to mix providers (OpenAI, Anthropic, Gemini, AWS, Azure) behind one OpenAI-format API
- you need weighted or round-robin load balancing across channels and keys
- you prefer no web dashboard and pure YAML configuration

## When to avoid
- you need a full commercial LLM gateway with a web UI, billing, and team management
- you only use a single provider and don't need unification
- you need features like fine-tuning management or model hosting

## Facets
- artifact type: service
- maturity: active
- function: api-gateway, llm-inference, proxy, middleware, configuration-management
- domain: large-language-models, artificial-intelligence, apis, self-hosted, backend
- platform: self-hosted, python, cross-platform
- tags: openai-compatible, llm-gateway, api-unification, load-balancing, channel-management, no-frontend, yaml-config, docker, web-server

## Member repositories
- yym68686/uni-api (main) score 86

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:04:09.357276+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-30T05:07:13.573562+00:00, confidence not recorded.
  - readme: https://github.com/yym68686/uni-api (fetched 2026-08-28T04:04:09.357276+00:00, sha bbd42c64c2d0)
  - homepage: https://0-0.pro/r/uniapi (fetched 2026-08-29T12:17:47.070836+00:00, sha bb02677479a4)
  - site_page: https://0-0.pro/docs/en (fetched 2026-08-29T12:17:47.080809+00:00, sha f06467ee8038)
- Data as of 2026-08-30T08:39:29.467469+00:00.