Skip to content

Perf: load weights, create KV cache, initialize tokenizer in parallel #3215

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom
Open
wants to merge 1 commit into
base: main
Choose a base branch
from
Open

Perf: load weights, create KV cache, initialize tokenizer in parallel #3215

wants to merge 1 commit into from

Conversation

@Bekaboo
Copy link

@Bekaboo Bekaboo commented Apr 27, 2025

Use multiple thread to load weights, cache and tokenizer, should slightly improve the initialization and TTFT time.

img_6
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

1 participant

@Bekaboo