upvote
We're serving it around 150-200tok/s (uses our new speculative decoding implementation on a DFlash2 draft model).

https://mixlayer.com, LAUNCH-Q38-27B gets you $5 in credits if you want to kick the tires.

reply
I tried in your playground and got 14.2 tok/s?
reply
apologies we just got a sudden burst of new users and traffic, it's scaling up now.
reply
just added 8 more H200s to the cluster, if you (or anyone else) runs into issues please feel free to drop me a message: zack at mixlayer.com
reply
Works much better now! Got 103.9 tok/s, not quite 200 - but still amazing! Thanks for sharing
reply
Something a lot of model providers don't talk about: any time an engine uses speculative decoding the throughput will depend on how much your output token distribution matches what the draft model was trained on.

The DFlash2 draft model we're using was trained on a lot of code, so if you use it in a coding agent you'll probably notice it run a lot faster (we've seen it break 300 tok/s).

reply
FYI, I might be missing something but I think your billing system might not be working well - I'm not seeing any indication in the UI that my usage is being deducted from the $5 of free credits.
reply
Hey Daniel! It's a bit hidden, but at the bottom of the billing page there's a "Credits" section which should show usage of any active credits and the balance remaining. The usage/billing metrics are batched/handled async so it might take a minute or so for usage to be reflected. Let us know if it feels off.
reply
I don't see any kind of input cache discount listed on your pricing page. Do you offer that, or is all input priced the same?
reply
deleted
reply
any way to see the tok/s for all the models listed on your homepage? curious which has the best speed/quality tradeoff for me
reply
This feels great
reply