English | 中文
A lightweight AI model status monitoring panel. Single-file deployment, zero external dependencies (except SQLite). Supports multi-group, multi-model monitoring with Uptime Kuma-compatible push. Compatible with NewAPI.
Web Dashboard
NewAPI Compatible
# Copy config template
cp config.template.json config.json
# Edit config.json with your models and API keys
# Run directly
go run .
# Visit http://localhost:8080
# Windows
$env:CGO_ENABLED="0"
go build -ldflags "-s -w" -o status-windows-amd64.exe .
# Linux cross-compile
$env:CGO_ENABLED="0"
$env:GOOS="linux"
$env:GOARCH="amd64"
go build -ldflags "-s -w" -o status-linux-amd64 .
Build output is a single executable + status.html. Just place them in the same directory to deploy.
Config file is config.json. See config.template.json for a template.
{
"port": 8080, // Listen port, default 8080, requires restart
"interval": 60, // Check interval in seconds, default 60
"frontend": { // Frontend display config
"title": "My API Status", // Page title (used for both <title> and <h1>)
"icon": "https://example.com/icon.png", // Favicon and logo, empty to hide
"subtitle": "Real-time service monitoring", // Subtitle below the heading
"footer": "© 2026 <a href=\"/\">Home</a>" // Footer, supports HTML
},
"kuma": { // Uptime Kuma push config (optional)
"baseURL": "https://status.example.com",
"slug": "my-status-page"
},
"groups": [ /* see below */ ]
}
{
"name": "OpenAI", // Top-level group name
"subGroups": [
{
"name": "GPT-4o", // Sub-group name (displayed as model name on page)
"kumaSlug": "gpt-4o", // Uptime Kuma independent push slug (optional)
"models": [
{
"id": "gpt-4o", // Model ID for API calls
"provider": "openai-compat", // Protocol type
"baseURL": "https://api.openai.com/v1",
"key": "sk-xxx", // API Key
"timeout": 30 // Timeout in seconds, default 30
}
]
}
]
}
| provider | Description | Endpoint |
|---|---|---|
openai-compat |
OpenAI-compatible protocol (default) | POST {baseURL}/chat/completions |
openai-response |
OpenAI Responses API | POST {baseURL}/responses |
google |
Google Gemini API | POST {baseURL}/models/{id}:generateContent |
anthropic |
Anthropic Claude API | POST {baseURL}/messages |
OpenAI-compatible (most proxy services):
{
"id": "gpt-4o-mini",
"provider": "openai-compat",
"baseURL": "https://api.example.com/v1",
"key": "sk-your-key",
"timeout": 30
}
Anthropic Claude:
{
"id": "claude-sonnet-4-20250514",
"provider": "anthropic",
"baseURL": "https://api.anthropic.com",
"key": "sk-ant-your-key",
"timeout": 30
}
Google Gemini:
{
"id": "gemini-2.5-flash",
"provider": "google",
"baseURL": "https://generativelanguage.googleapis.com/v1beta",
"key": "your-api-key",
"timeout": 30
}
OpenAI Responses API (Codex):
{
"id": "gpt-5.4-mini",
"provider": "openai-response",
"baseURL": "https://api.openai.com/v1",
"key": "sk-your-key",
"timeout": 30
}
| Endpoint | Description |
|---|---|
GET / |
Status dashboard page |
GET /api/status |
Status data (polled by frontend) |
GET /api/config |
Config info (interval + frontend) |
GET /api/status-page/{slug} |
Uptime Kuma-compatible status page |
GET /api/status-page/heartbeat/{slug} |
Uptime Kuma-compatible heartbeat data |
{
"title": "My API Status",
"overall": { "status": "up", "label": "Running" },
"last_completed_at": "2026-06-07T12:00:00Z",
"checking": false,
"sections": [
{
"name": "OpenAI",
"rows": [
{
"name": "gpt-4o",
"sub_name": "GPT-4o",
"status": "up",
"label": "Normal",
"health_score": 99.5,
"latency_ms": 230,
"last_checked_at": "2026-06-07T12:00:00Z",
"history": [{ "s": "up", "v": 1.0 }]
}
]
}
]
}
status values: up / down / degraded / unknown
{
"interval": 60,
"frontend": {
"title": "My API Status",
"icon": "https://example.com/icon.png",
"subtitle": "...",
"footer": "..."
}
}
Two push modes supported:
- Global push: Set
kuma.baseURLandkuma.slugto push the average status of all models to one slug - Per-subgroup push: Set
kumaSlugin each subGroup for independent push per group
Also compatible with NewAPI and other tools using the Uptime Kuma public status page protocol, providing /api/status-page/ and /api/status-page/heartbeat/ endpoints.
The frontend config section controls all branding elements without modifying HTML:
title: Browser tab title + page main headingicon: Favicon + page logo (hidden when empty)subtitle: Text below the headingfooter: Footer content, supports inline HTML
- Go (single file,
modernc.org/sqliteCGO-free SQLite driver) - Pure HTML/CSS/JS, no frontend framework dependencies
- Storage: local SQLite (
status.db)

