---
date: 2025-11-06
title: Annotation Support in Experiment Compare View
description: Add human annotations while reviewing experiment results side-by-side. Review experiment outputs, assign scores, and leave comments while viewing full experiment context.
author: Marlies
ogImage: /images/changelog/2025-10-23-annotate-compare-view-overview.png
---

> **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 Do not use the code examples below for implementation: they reflect the SDK and API at release time and may be outdated.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 current documentation (https://langfuse.com/docs) and the API/SDK reference (https://api.reference.langfuse.com).

You can now annotate traces directly from the experiment compare view, streamlining the workflow of running experiments and adding human feedback.

**What's New**

Human annotation is now available directly in the experiment compare view. Select any cell to open the annotation side panel, where you can assign scores and leave comments while maintaining full experiment context. Use resulting annotation score data to compare experiment results across different prompt versions, model configurations or iterations of your application.

**How It Works**

  ![Annotate from compare view](/images/changelog/2025-10-23-annotate-compare-view-overview.png)

1. **Run an experiment**: Execute an [experiment via UI](/docs/evaluation/experiments/experiments-via-ui) or [SDK](/docs/evaluation/experiments/experiments-via-sdk) to test prompt versions or models
2. **Open compare view**: Navigate to the experiment comparison to review results side-by-side
3. **Select a cell**: Click any experiment item to open the annotation panel
4. **Add scores**: Assign values for the [score configs](/faq/all/manage-score-configs) you've set up
5. **Leave comments**: Add context for your team about specific outputs
6. **Move to next item**: Click "Annotate" on the next cell you'd like to review

The UI updates optimistically as you add scores, providing immediate feedback while data persists in the background. Summary metrics in the compare view reflect your annotations as you work.

**Score Configurations**

Before annotating, [create score configs](/faq/all/manage-score-configs) to standardize scoring criteria across your team. Score configs support:

- **Numerical scores**: Define min/max ranges for quantitative evaluation
- **Categorical scores**: Set custom categories for classification tasks
- **Binary scores**: Use boolean values for pass/fail judgments

Standardized score configs ensure consistency and make results comparable across experiments and team members. You can update score configs at any time.

**Getting Started**

1. Set up [score configs](/faq/all/manage-score-configs) for your evaluation criteria
2. Run an [experiment via UI](/docs/evaluation/experiments/experiments-via-ui)
3. Open the compare view and start annotating

## Learn More

- [Human Annotation Documentation](/docs/evaluation/evaluation-methods/annotation)
- [Score Config Management](/faq/all/manage-score-configs)
- [Experiments via UI](/docs/evaluation/experiments/experiments-via-ui)
- [Experiments via SDK](/docs/evaluation/experiments/experiments-via-sdk)

<!-- 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/2025-10-23-annotate-from-compare-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>.