olive gate
https://huggingface.co/collections/ai21labs/jamba-17-68653e9be386dc69b1f30828
Jamba Mini 1.7 offers new improvements to our Jamba open model family. This new version builds on the novel SSM-Transformer hybrid architecture, 256K context window, and efficiency gains of previous versions, while introducing improvements in grounding and instruction-following.
marble imp
I've heard of the Jamba models before but never used them, are they any good?
plush galleon
gg
trim bloom
I've heard of the Jamba models before but never used them, are they any good?
no
jamba has been garbage
not one single use case that it is good got
timber hornet
Haven't heard of the IFEcal benchmark... sounds a bit shitty... 🤣
keen lily
I'm actually enjoying this model, most of the time they're all garbage but this one is pretty nice 
ancient saddle
still doesn't support basic response format (such as structured table). I noted this on 1.6: #1347257225860878379 message )
ancient saddle
Tested Jamba 1.7 (Mini & Large):
Still don't support basic response format (such as structured table).
Strangely, the older 1.5 models performed better.
API pricing is outdated.
They feel and perform absolutely ancient, as if we traveled back in time 1.5+ years ago.
Consequently, I see no use case for this family. YMMV.
olive gate
Lol