Roger Wang
Roger Wang
> Hi~ Does vllm support multiple image input now? @xyfZzz Not yet - This PR itself allows profiling with multiple image input but there are still a few things we...
@ekagra-ranjan I've merged this branch from main so let's see if it helps!
This is an interesting bug, but I don't think adding `output.text` check here is the right way to fix it. If anything, we should probably build a check around token_ids....
Closing as fixed by https://github.com/vllm-project/vllm/pull/9549
@Isotr0py Sorry for the late review! Overall LGTM but let me update this PR with main and test it out locally with a better GPU and see if the tests...
Do you want to remove the skip decorator and see if it actually passes on our CI?
Let’s skip the TP addition for a later PR then
Thank you for making the contribution! I’ll take a look tonight!
Hello! @mrsalehi and @sangho-vision. Friendly ping to see when we can get an update on this PR. If you need any help please let us know!
@sangho-vision Sorry for the delay - I will review this PR either tonight or tomorrow!