> ## Documentation Index
> Fetch the complete documentation index at: https://docs.budecosystem.com/llms.txt
> Use this file to discover all available pages before exploring further.

# Introduction to Dashboard

> Monitor platform health, usage, and model quality from a single overview

## Overview

The Dashboard module gives teams a single operational view of Bud AI Foundry. It combines request activity, model inventory, deployment readiness, cluster posture, and model quality signals so you can detect issues early and make faster routing or scaling decisions.

```mermaid theme={null}
graph TD
    A[Dashboard Home] --> B[Global Summary Cards]
    A --> C[Operational Charts]
    B --> D[Requests and Model Inventory]
    B --> E[Endpoints Clusters Projects]
    C --> F[API Calls Latency Throughput]
    C --> G[Model Usage Accuracy Tokens]
```

## Key Features

**Operational Summary Cards**
Track total requests, total models, running endpoints, inactive clusters, and project member footprint at a glance.

**Built-in Time Windows**
Each major chart supports **Last 24 Hrs**, **Last 7 Days**, and **Last 30 Days** so you can compare short-term anomalies vs. longer trends.

**Top-Model Intelligence**
See top models by usage, benchmark accuracy, and input/output token mix to balance cost, quality, and performance.

**Project-Level Visibility**
API calls, latency, and throughput views are grouped by project to help you identify hotspots quickly.

## What You Can Monitor

| Area               | What you see                                | Why it matters                           |
| ------------------ | ------------------------------------------- | ---------------------------------------- |
| **Requests**       | 7-day request total, trend sparkline, delta | Detect sudden usage spikes or drops      |
| **Models**         | Total model count with cloud/local split    | Track inventory growth and hosting mix   |
| **Readiness**      | Endpoint, cluster, and project status cards | Spot infra risk before incidents         |
| **Performance**    | API calls, latency, throughput charts       | Prioritize reliability and capacity work |
| **Quality & Cost** | Accuracy leaderboard + token metrics        | Balance model quality and token spend    |

## How It Fits in Daily Operations

1. **Start in Dashboard** to identify where trends changed.
2. **Drill down** to Models, Clusters, Projects, or Observability.
3. **Take action** (scale, reroute, investigate, or evaluate).
4. **Return to Dashboard** to confirm outcomes.

## Getting Started

<CardGroup cols={3}>
  <Card title="Quick Start" icon="play" href="/dashboard/quickstart">
    Learn core navigation and chart filters
  </Card>

  <Card title="Dashboard Concepts" icon="book" href="/dashboard/dashboard-concepts">
    Understand cards, deltas, and ranking logic
  </Card>

  <Card title="First Monitoring Workflow" icon="graduation-cap" href="/dashboard/creating-first-dashboard">
    Build a repeatable operations review
  </Card>
</CardGroup>