Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Multimodal Video Inference on LLaVA Video Sequence 1.6
Loading...
2.2
Compression Ratio (×)
Sali-Cache
0.952
1.276
1.6
1.924
Feb 15, 2026
Compression Ratio (×)
Peak VRAM (GB)
Cache Size (GB)
Updated 5mo ago
Evaluation Results
Method
Method
Links
Compression Ratio (×)
Peak VRAM (GB)
Cache Size (GB)
Sali-Cache
Backbone=Mistral 7B, P...
2026.02
2.2
-
-
H2O
Backbone=Mistral 7B, P...
2026.02
1.05
-
-
Baseline
Backbone=Mistral 7B, P...
2026.02
1
-
-
Feedback
Search any
task
Search any
task