Dflash is much slower when enabling. #816
Unanswered
justin1230
asked this question in
Q&A
Replies: 0 comments
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Uh oh!
There was an error while loading. Please reload this page.
-
Hi,
I just updated to version 0.3.6 oMLX and wanna try the new dflash.
I am using Qwen3.5-9B-MLX-4bit, with the Qwen3.5-9B-DFlash draft model.
Enabled the dflash and selected the draft model. But the speed is much slower with just simple prompt in testing. Am i missing anything?
Beta Was this translation helpful? Give feedback.
All reactions