> ## Documentation Index
> Fetch the complete documentation index at: https://docs.cloosphere.ai/llms.txt
> Use this file to discover all available pages before exploring further.
# Evaluation Overview
> Measure quality with Feedback, Arena & Leaderboard, and Auto-Evaluation, then diagnose root causes with Tracing
Admin › Evaluation
The evaluation feature measures AI response quality in three ways: **Feedback (manual) · Arena & Leaderboard · Auto-Evaluation**.
* By combining direct user feedback, blind model comparison, and LLM-based automatic scoring, you can build a systematic quality-management framework.
* When a measurement surfaces a problem, use **Tracing** to step through how that request was processed and diagnose the root cause.
```mermaid theme={null}
flowchart TB
subgraph Manual_Evaluation
A[Check response] --> B[Like / Dislike]
B --> C[Enter comment]
C --> D[Feedback data]
end
subgraph Arena_Leaderboard
K[Blind-compare two models] --> L[User picks]
L --> M[Compute Elo ranking]
end
subgraph Auto_Evaluation
E[Agent response] --> F[Sampling]
F --> G[Judge LLM evaluates]
G --> H[Record score]
end
D --> I[Evaluation dashboard]
M --> I
H --> I
I --> N[Tracing · root-cause diagnosis]
N --> J[Quality improvement]
```
***
## Evaluation Methods
Three methods **measure** quality, and Tracing **diagnoses** the root cause of any problem they surface.
* Measurement tells you what is bad; tracing tells you why.
| Category | Method | Signal | When to use |
| ------------ | ------------------------------------------------------- | ----------------------- | ------------------------------------------ |
| **Measure** | [Feedback](/en/monitoring/feedback) | Like/Dislike · comments | Collect everyday quality signals |
| **Measure** | [Arena & Leaderboard](/en/monitoring/arena-leaderboard) | Blind choice → Elo | Compare models head-to-head |
| **Measure** | [Auto-Evaluation](/en/monitoring/auto-evaluations) | Judge-LLM score (0–1) | Large-scale, continuous quality monitoring |
| **Diagnose** | [Tracing](/en/monitoring/tracing) | Processing trace | Analyze the cause of low scores |
***
## Detailed Guides
Jump to the page that fits your goal.
Collect, view, and manage likes, dislikes, and comments
Blind comparison · Elo-based model ranking
Judge-LLM automatic scoring — enable, results, statistics, export
Trace the cause of low evaluation scores