Skip to content

Commit 4ea796c

Browse files
authored
Merge pull request #3406 from Agenta-AI/docs/changelog-playground-ux-improvements
docs(changelog): add Playground UX improvements announcement
2 parents 6be4cf6 + c3cae61 commit 4ea796c

File tree

3 files changed

+65
-0
lines changed

3 files changed

+65
-0
lines changed
Lines changed: 49 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,49 @@
1+
---
2+
title: "Playground UX Improvements"
3+
slug: playground-ux-improvements-jan-2026
4+
date: 2026-01-13
5+
tags: [v0.73.0]
6+
description: "See provider costs upfront, run evaluations directly from the Playground, and collapse test cases for easier navigation."
7+
---
8+
9+
# Playground UX Improvements
10+
11+
<div style={{display: 'flex', justifyContent: 'center', marginTop: "20px", marginBottom: "20px", flexDirection: 'column', alignItems: 'center'}}>
12+
<iframe
13+
width="100%"
14+
height="500"
15+
src="https://www.youtube.com/embed/JqVj-gsnSgk"
16+
title="Playground UX Improvements"
17+
frameBorder="0"
18+
allow="accelerometer; autoplay; clipboard-write; encrypted-media; gyroscope; picture-in-picture; web-share"
19+
allowFullScreen
20+
></iframe>
21+
</div>
22+
23+
## Overview
24+
25+
This release brings three quality-of-life improvements to the Playground that make testing and iterating on prompts faster and more convenient.
26+
27+
## What's New
28+
29+
### Provider Cost Display
30+
31+
You can now see the cost per million tokens directly in the provider selection dropdown. This helps you make informed decisions about which model to use based on both capability and cost. No more switching to external pricing pages to compare costs.
32+
33+
### Run Evaluations from the Playground
34+
35+
You can now trigger evaluations directly from the Playground without navigating to the evaluation menu. When you're testing a prompt and want to run a full evaluation, click the evaluate button to start an evaluation run with your current configuration. This keeps you in flow when iterating on prompts.
36+
37+
### Collapsible Test Cases
38+
39+
Test cases in the Playground can now be collapsed. This is especially useful when working with large test sets or test cases with long inputs and outputs. Collapse completed test cases to focus on what you're working on. You can still see a preview of each test case to maintain context while navigating through your data.
40+
41+
42+
## Getting Started
43+
44+
These features are available now in the Playground. Start using them today:
45+
46+
- Open the Playground and check the provider dropdown to see costs
47+
- Click the evaluate button to run evaluations directly
48+
- Use the collapse controls on test cases to manage your view
49+

docs/blog/main.mdx

Lines changed: 10 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -12,6 +12,16 @@ import Image from "@theme/IdealImage";
1212

1313

1414

15+
### [Playground UX Improvements](/changelog/playground-ux-improvements-jan-2026)
16+
17+
_13 January 2026_
18+
19+
**v0.73.0**
20+
21+
Three quality-of-life improvements to the Playground: You can now see provider costs per million tokens directly in the model selection dropdown. You can run evaluations directly from the Playground without navigating to the evaluation menu. And you can collapse test cases to navigate large test sets more easily.
22+
23+
---
24+
1525
### [Chat Sessions in Observability](/changelog/chat-sessions-observability)
1626

1727
_9 January 2026_

web/oss/src/components/SidebarBanners/data/changelog.json

Lines changed: 6 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -1,4 +1,10 @@
11
[
2+
{
3+
"id": "changelog-2026-01-13-playground-ux",
4+
"title": "Playground UX Improvements",
5+
"description": "See provider costs, run evaluations directly, and collapse test cases.",
6+
"link": "https://agenta.ai/docs/changelog/playground-ux-improvements-jan-2026"
7+
},
28
{
39
"id": "changelog-2026-01-09-chat-sessions",
410
"title": "Chat Sessions in Observability",

0 commit comments

Comments
 (0)