# wyeeeee/hajimi

这是一个基于 FastAPI 构建的 Gemini API 代理

Repository: https://github.com/wyeeeee/hajimi
Canonical: https://ross.abutalabs.com/products/hajimi
Language: Python
License: NOASSERTION
License Family: other
Archived: true
Last push: 2025-09-21T02:03:39+00:00

## Health v2 (maintenance only)
Score: 10/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 43, release rhythm 40, longevity 37
- inputs: {"age_days": 520, "days_push": 347, "days_rel": 454, "gap_med": 1, "n_releases_24m": 20}
- flags: archived, no_license
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 1727, forks 5336 (observed 2026-08-28T04:05:28.302558+00:00)

## What it is
Hajimi is a FastAPI-based proxy for Google's Gemini API that exposes OpenAI-compatible endpoints like /v1/chat/completions. It provides API key pooling and rotation, rate limiting, optional password protection, fake streaming, concurrent request caching, and web search mode, deployable on Hugging Face Spaces, Claw Cloud, Render, or locally.

## Use cases
- use gemini models through openai-compatible clients like sillytavern
- proxy gemini api with multiple api keys and rotation
- self-host an openai-format endpoint for gemini
- deploy a gemini proxy on huggingface spaces for free
- add rate limiting and password protection to a gemini proxy
- enable gemini web search via model name suffix
- convert openai chat requests to gemini format

## When to choose
- you need an OpenAI-compatible gateway in front of Gemini or Vertex AI
- you want key pooling, rate limits, and a web UI for configuration
- you want free one-click deployment on Hugging Face Spaces or Claw Cloud

## When to avoid
- you need a fully supported, production-hardened gateway with a clear license (license is unspecified)
- you only use OpenAI or Anthropic models and never Gemini
- you require strict compliance, since the proxy includes request-disguise features that may violate provider terms

## Facets
- artifact type: service
- maturity: active
- function: api-gateway, proxy, http-server, middleware, rate-limiting, chatbot
- domain: large-language-models, apis, self-hosted, backend, web-development
- platform: python, self-hosted, cross-platform
- tags: gemini-api-proxy, openai-compatible, fastapi, api-key-pooling, huggingface-spaces, vertex-ai, docker, web-server

## Member repositories
- wyeeeee/hajimi (main) score 10

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:05:28.302558+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 2026-08-30T03:32:30.028080+00:00, confidence not recorded.
  - readme: https://github.com/wyeeeee/hajimi (fetched 2026-08-28T04:05:28.302558+00:00, sha 6eedecaf60ca)
- Data as of 2026-08-30T08:39:29.467469+00:00.