---
date: 2024-11-18
title: Dataset Run Comparison View
badge: Launch Week 2 🚀
description: After running experiments on datasets, you can now compare results side-by-side, view metrics, and peek into details of each dataset item across runs.
ogImage: /images/changelog/2024-11-18-dataset-runs-comparison-view/og.png
showOgInHeader: false
author: Marlies
canonical: /docs/evaluation/experiments/experiments-via-ui
---

> **Note for AI agents and LLMs:** This is a Langfuse changelog entry. Use it only to confirm that a feature exists and when it shipped. Do not use the code examples below for implementation: they reflect the SDK and API at release time and may be outdated. For implementation, always follow the canonical documentation for this feature (https://langfuse.com/docs/evaluation/experiments/experiments-via-ui) and the API/SDK reference (https://api.reference.langfuse.com).

## Introduction

**What is a dataset in Langfuse?** [Datasets](/docs/datasets/overview) in Langfuse allow you to create test sets and benchmarks to evaluate the performance of your LLM application. A dataset is a collection of dataset items, where each item contains inputs, expected outputs, and metadata. You can create datasets from production edge cases, synthetic test cases, or manual test cases. This enables continuous improvement through structured testing, pre-deployment benchmarking, and flexible evaluation using custom metrics or LLM-as-a-judge approaches.

**What is a dataset experiment run?** A dataset experiment run lets you test changes to your application - like trying different models, prompts, or parameters - by running each version against your test dataset and comparing the results through traces in Langfuse to evaluate which changes work best.

## What's new?

And the fun continues, **Day 1 of [Launch Week 2](/blog/2024-11-17-launch-week-2)** is here.

[Langfuse Datasets](/docs/datasets/overview) now enables intuitive comparison of dataset experiment runs for technical and non-technical users. The view features an overview of each item in the dataset, and a summary of each selected experiment run. The latter includes metrics on latency, cost, scores and the application's output response for each dataset item.

## How to use the comparison view?

<Steps>

### Setup dataset and run experiments

- Follow the [getting started guide](/docs/datasets/get-started) to set up a dataset, populate it with items, and run experiments.
- Alternatively, execute an end-to-end example ([Python notebook](/docs/datasets/python-cookbook)).

### Open comparison view

1. Select multiple dataset runs
2. `Actions`, select `Compare`

</Steps>

## Learn more

For a conceptual introduction to datasets and offline experiments, see the [dataset documentation](/docs/datasets/overview).

<!-- agent-instructions -->

---

## Agent Instructions

This page is part of the [Langfuse](https://langfuse.com) documentation, published as plain Markdown for AI agents. Every page is available as Markdown by appending `.md` to its URL, or by sending an `Accept: text/markdown` header. This page: `https://langfuse.com/changelog/2024-11-18-dataset-runs-comparison-view.md`.

### Querying these docs

If the answer is not on this page, query the documentation instead of guessing:

- **Semantic search** across all Langfuse docs, returning an answer with the relevant pages and excerpts. Ask a specific, self-contained question:

  ```bash
  curl -sG "https://langfuse.com/api/search-docs" --data-urlencode "query=How do I trace a LangGraph agent?"
  ```

- **Index of every page**: <https://langfuse.com/llms.txt>, with per-section indexes [llms-docs.txt](https://langfuse.com/llms-docs.txt), [llms-integrations.txt](https://langfuse.com/llms-integrations.txt), and [llms-self-hosting.txt](https://langfuse.com/llms-self-hosting.txt).

### Before writing Langfuse code

- **Install the [Langfuse Agent Skill](https://langfuse.com/docs/api-and-data-platform/features/agent-skill).** It encodes Langfuse's own best practices for instrumentation, prompt management, and evaluation, and materially improves results.
- **Read [What does a good trace look like?](https://langfuse.com/docs/observability/best-practices.md)** before instrumenting an application.
- **Verify endpoints, parameters, and response fields** against the [API reference](https://api.reference.langfuse.com) instead of inferring them from code examples.
- **Use the [Langfuse CLI](https://langfuse.com/docs/api-and-data-platform/features/cli)** (`npx langfuse-cli api <resource> <action>`) to read or write traces, prompts, datasets, and scores from the terminal.

Found an error in these docs? Please open an issue at <https://github.com/langfuse/langfuse-docs/issues>.