Lightweight and fast vision model, does a decent job captioning photos.
vision
1,999 Pulls Updated 8 months ago
Updated 8 months ago
8 months ago
d614b616cc22 · 6.0GB
model
archllama
·
parameters2.7B
·
quantizationF16
5.4GB
projector
archclip
·
parameters320M
·
quantizationF16
640MB
template
Q: {{ .Prompt }}
A:
19B
params
{"stop":["Q:","A:"],"temperature":0}
37B