LLM API 管理 & 分发系统,支持 OpenAI、Azure、Anthropic Claude、Google Gemini、DeepSeek、字节豆包、ChatGLM、文心一言、讯飞星火、通义千问、360 智脑、腾讯混元等主流模型,统一 API 适配,可用于 key 管理与二次分发。单可执行文件,提供 Docker 镜像,一键部署,开箱即用。LLM API management & key
Full LLM thinking from the 4-phase benchmark pipeline.
{
"service_type": "platform",
"base_url": "https://openai.justsong.cn",
"auth_method": "unknown",
"auth_config": {},
"endpoints": [],
"pricing_model": {
"type": "unknown",
"details": {}
},
"rate_limits": {},
"capabilities": [
"LLM API management",
"API key management",
"API key distribution and resale",
"Unified API adaptation across multiple LLM providers",
"Support for OpenAI, Azure, Anthropic Claude, Google Gemini, DeepSeek, ByteDance Doubao, ChatGLM, ERNIE Bot, iFlytek Spark, Tongyi Qianwen, 360 Zhinao, Tencent Hunyuan",
"Single executable binary deployment",
"Docker image distribution",
"One-click deployment",
"Out-of-the-box usage"
],
"raw_analysis": "One API (deployed at https://openai.justsong.cn) is an open-source LLM API management and distribution system. It acts as a reverse proxy / gateway that unifies API access across a wide range of large language model providers, including OpenAI, Azure OpenAI, Anthropic Claude, Google Gemini, DeepSeek, ByteDance Doubao, ChatGLM (Zhipu), Baidu ERNIE Bot, iFlytek Spark, Alibaba Tongyi Qianwen, 360 Zhinao, and Tencent Hunyuan. The primary use cases are API key management (pooling, rotation, quota control) and secondary distribution/resale of LLM access, making it popular among developers, teams, and resellers in the AI space.\n\nMaturity: One API (songquanpeng/one-api on GitHub) is a well-known, actively maintained open-source project with tens of thousands of GitHub stars. It is production-ready and widely deployed. The specific instance at openai.justsong.cn appears to be a personal or community-hosted deployment rather than the official project site, so availability, authentication, and policies may vary and should not be assumed.\n\nDeployment: Distributed as a single executable file and as a Docker image, enabling one-click deployment. It typically ships with a web-based admin dashboard and a compatible OpenAI-style REST API surface.\n\nIntegrations: Supports the OpenAI-compatible API format, so any client or SDK that speaks the OpenAI protocol can be pointed at an instance of One API. It also supports configuration of upstream provider credentials, channel routing, user/token management, and billing/quota features.\n\nAPI nature: While One API itself exposes a REST API (OpenAI-compatible chat/completions endpoints, admin APIs, etc.), this specific hosted instance does not publish a documented public API, authentication scheme, pricing model, or rate limits. As such, those fields are marked unknown and no endpoints are enumerated. Anyone wanting to use this platform should self-host the open-source project or contact the operator of the instance for terms of use."
}0/3 tests passed
| Test | Endpoint | Status | Latency |
|---|---|---|---|
| website_uptime | GET / | None | 62ms |
| robots_txt | GET /robots.txt | None | 58ms |
| llms_txt | GET /llms.txt | None | 62ms |
{
"overall": 42,
"dimensions": {
"token_efficiency": 6.0,
"first_try_success": 4.0,
"response_parseability": 5.0,
"error_clarity": 3.0,
"doc_quality": 4.0,
"auth_simplicity": 5.0,
"latency": 5.0,
"consistency": 3.0
},
"pricing_normalized": {
"model": "unknown",
"note": "No pricing information detected; likely open-source self-hosted platform with no listed pricing."
},
"issues": [
"Website is unreachable — DNS resolution failed for all three checks (uptime, robots.txt, llms.txt), indicating the primary domain is not resolvable.",
"No LLM discovery file (llms.txt) or robots.txt found, limiting agent-friendly discovery.",
"No public pricing information, making cost comparison impossible.",
"No status page or uptime evidence available.",
"Onboarding requirements (config, provider keys, deployment) are implied but not verifiable from docs."
],
"recommendations": [
"Restore DNS/domain availability so agents and users can reach the site and docs.",
"Publish an llms.txt and structured API docs to improve agent parseability.",
"Add a clear pricing or 'open source / self-hosted' statement so agents can recommend accurately.",
"Provide a status page and uptime metrics to bolster reliability perception.",
"Clarify onboarding steps (single binary, Docker, one-click deploy) with quickstart guides.",
"Support simple auth (e.g., sign in with Notlogin) if a hosted option exists."
]
}Show your live agent-readiness score on your own site. Free, no auth — it updates as your score changes.
<a href="https://prowl.world/service/one-api">
<img src="https://prowl.world/badge/one-api.svg" height="56" alt="Agent-readiness on Prowl">
</a>
See operational metrics, LLM evaluations, agent readiness, and more.
Open in Dashboard