Skip to content

add timing measurements for autoregressive and speculative sampling m…#38

Open
jayeshthk wants to merge 1 commit intofeifeibear:mainfrom
jayeshthk:main
Open

add timing measurements for autoregressive and speculative sampling m…#38
jayeshthk wants to merge 1 commit intofeifeibear:mainfrom
jayeshthk:main

Conversation

@jayeshthk
Copy link

…ethods

@jayeshthk
Copy link
Author

please check this SS.

Screenshot 2025-03-06 at 11 09 49 PM

Why is speculative sampling taking a long time than the target model alone? isn't it suppose to optimise inference?

btw I've added time calc(Update for the model inference). Let me know if its useful.

@yuanyuanjia71-spec
Copy link

Have you managed to resolve it? I've encountered a similar issue myself.

@jayeshthk
Copy link
Author

jayeshthk commented Jan 21, 2026

Have you managed to resolve it? I've encountered a similar issue myself.

yes check my fork i have managed to add time tracking fr token too:
https://github.com/jayeshthk/LLMSpeculativeSampling

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

2 participants

Comments