webmarker-js 0.0.7 → 0.1.0
Sign up to get free protection for your applications and to get access to all the features.
- package/README.md +3 -2
- package/package.json +1 -1
- package/test-results/.last-run.json +4 -2
package/README.md
CHANGED
@@ -11,10 +11,11 @@ Mark web pages for use with vision-language models.
|
|
11
11
|
|
12
12
|
## Overview
|
13
13
|
|
14
|
-
🚧 Under Construction
|
15
|
-
|
16
14
|
**WebMarker** adds visual markings with labels to elements on a web page. This can be used for [Set-of-Mark (SoM)](https://github.com/microsoft/SoM) prompting, which improves visual grounding abilities of vision-language models such as GPT-4o, Claude 3.5, and Google Gemini 1.5.
|
17
15
|
|
16
|
+
![Screenshot of marked Google homepage](https://github.com/user-attachments/assets/722e1034-06d4-4ccd-a7d6-f03749435681)
|
17
|
+
|
18
|
+
|
18
19
|
## How it works
|
19
20
|
|
20
21
|
**1. Call the `mark()` function**
|
package/package.json
CHANGED