Kwai-Keye
/

Keye-VL-8B-Preview

Video-Text-to-Text

feature-extraction

Model card Files Files and versions

Kwai-Keye commited on Jun 27

Commit

a88b27e

·

verified ·

1 Parent(s): 0046c54

Update README.md

Files changed (1) hide show

README.md +1 -1

README.md CHANGED Viewed

@@ -465,7 +465,7 @@ The post-training phase of Kwai Keye is meticulously designed into two phases wi
 ## 📈 Experimental Results
-![image](https://github.com/user-attachments/assets/a27cc0b8-e511-4879-969a-b6bc90f61c7e)
 1. Keye-VL-8B establishes itself with powerful, state-of-the-art perceptual abilities that are competitive with leading models.
 2. Keye-VL-8B demonstrates exceptional proficiency in video understanding. Across a comprehensive suite of authoritative public video benchmarks, including Video-MME, Video-MMMU, TempCompass, LongVideoBench, and MMVU, the model's performance significantly surpasses that of other top-tier models of a comparable size.

 ## 📈 Experimental Results
+![image](https://github.com/user-attachments/assets/08edb828-5532-465c-9714-e27b8dacdbb9)
 1. Keye-VL-8B establishes itself with powerful, state-of-the-art perceptual abilities that are competitive with leading models.
 2. Keye-VL-8B demonstrates exceptional proficiency in video understanding. Across a comprehensive suite of authoritative public video benchmarks, including Video-MME, Video-MMMU, TempCompass, LongVideoBench, and MMVU, the model's performance significantly surpasses that of other top-tier models of a comparable size.