# Evaluate and trace LLM apps

Capability · https://godooo.ai/en/capability/llm-evaluation

> Evaluate and trace LLM apps: 6 AI tools in this graph can do it, including Langfuse, promptfoo, Phoenix, Ragas and DeepEval.

Trace what an LLM app does and score its answers against test sets.

- **Takes in:** Text, Data
- **Gives out:** Data

## Tools that do it

| Tool | Type | How well | Pricing | Open source | How you run it |
| --- | --- | --- | --- | --- | --- |
| [Langfuse](https://godooo.ai/en/product/langfuse) | Product | main job* | Open source, free | Yes | Web, Self-host, API |
| [promptfoo](https://godooo.ai/en/cli/promptfoo) | CLI | main job* | Open source, free | Yes | Terminal, Runs offline, Self-host |
| [Phoenix](https://godooo.ai/en/framework/phoenix) | Framework | main job* |  | Yes | Self-host, Runs offline, API, Terminal |
| [Ragas](https://godooo.ai/en/framework/ragas) | Framework | main job* |  | Yes | Runs offline, Self-host |
| [DeepEval](https://godooo.ai/en/framework/deepeval) | Framework | main job* |  | Yes | Self-host, Terminal, API |
| [Pydantic AI](https://godooo.ai/en/framework/pydantic-ai) | Framework | supported* |  | Yes | Terminal, API, Self-host, Runs offline |

\* proposed by a machine, awaiting calibration

## Related

- **Needed to:** [Build an AI agent](https://godooo.ai/en/goal/build-ai-agent), [Fine-tune a model](https://godooo.ai/en/goal/fine-tune-model)

## Questions

### What is Evaluate and trace LLM apps?

Trace what an LLM app does and score its answers against test sets. Evaluate and trace LLM apps: 6 AI tools in this graph can do it, including Langfuse, promptfoo, Phoenix, Ragas and DeepEval.

### Which tools are built for “Evaluate and trace LLM apps”?

Langfuse, promptfoo, Phoenix, Ragas and DeepEval.

### Are there free tools for “Evaluate and trace LLM apps”?

Langfuse, promptfoo, Phoenix, Ragas, DeepEval and Pydantic AI.

### Which open-source tools can do “Evaluate and trace LLM apps”?

Langfuse, promptfoo, Phoenix, Ragas, DeepEval and Pydantic AI.

### Which tools for “Evaluate and trace LLM apps” run offline?

promptfoo, Phoenix, Ragas and Pydantic AI.

---

Last updated 2026-09-24 · Markdown version of https://godooo.ai/en/capability/llm-evaluation. Every page on this site has one: add .md to its address.