text-generation-inference
feat: improve qwen2-vl startup
#2802
Merged

feat: improve qwen2-vl startup #2802

drbh merged 5 commits into main from improve-qwen2-vl-warmup
drbh
drbh drbh force pushed from 066addd7 to 60b9c187 1 year ago
Narsil
drbh drbh force pushed from 60b9c187 to 3cc82978 1 year ago
drbh drbh force pushed from 32a95640 to a3049f10 1 year ago
drbh drbh requested a review from Narsil Narsil 1 year ago
drbh drbh force pushed from a3049f10 to d671f6e0 1 year ago
drbh drbh changed the title feat: tokenize each request individually and increase warmup image size feat: improve qwen2-vl startup 1 year ago
drbh drbh force pushed from d671f6e0 to 320b520d 1 year ago
drbh feat: tokenize each request individually and increase warmup image size
1bcfba30
drbh feat: adjust rotary embed and avoid cuda graphs of size 2 and smaller
45e5c2c2
drbh fix: address image resize and rebase changes
822bd045
drbh feat: update to run qwen2-vl tests
bd59f961
drbh drbh force pushed from 35b528e8 to bd59f961 1 year ago
drbh fix: tweak param types
37f92f2c
drbh
drbh drbh merged eecca271 into main 1 year ago
drbh drbh deleted the improve-qwen2-vl-warmup branch 1 year ago

Login to write a write a comment.

Login via GitHub

Reviewers
Assignees
No one assigned
Labels
Milestone