GitLab

[Core] Fix tracking of model forward time to the span traces in case of PP>1 (#7440)

Signed-off-by: Rui Qiao <ruisearch42@gmail.com>

[Core] Subclass ModelRunner to support cross-attention & encoder sequences (towards eventual encoder/decoder model support) (#4942)
Co-authored-by: Andrew Feldman <afeld2012@gmail.com>
Co-authored-by: Nick Hill <nickhill@us.ibm.com>

Co-authored-by: Woosuk Kwon <woosuk.kwon@berkeley.edu>

Co-authored-by: Mor Zusman <morz@ai21.com>

Co-authored-by: Swapnil Parekh <swapnilp@ibm.com>
Co-authored-by: Joe G <joseph.granados@h2o.ai>
Co-authored-by: Antoni Baum <antoni.baum@protonmail.com>

Signed-off-by: Muralidhar Andoorveedu <muralidhar.andoorveedu@centml.ai>
Co-authored-by: Erez Schwartz <erezs@ai21.com>
Co-authored-by: Mor Zusman <morz@ai21.com>
Co-authored-by: tomeras91 <57313761+tomeras91@users.noreply.github.com>
Co-authored-by: Tomer Asida <tomera@ai21.com>
Co-authored-by: Zhuohan Li <zhuohan123@gmail.com>
Co-authored-by: Muralidhar Andoorveedu <muralidhar.andoorveedu@centml.ai>

Signed-off-by: Muralidhar Andoorveedu <muralidhar.andoorveedu@centml.ai>

Currently we need to call rotary embedding kernel for each LoRA, which makes it hard to serve multiple long context length LoRA. Add batched rotary embedding kernel and pipe it through.

It replaces the rotary embedding layer to the one that is aware of multiple cos-sin-cache per scaling factors.

Follow up of https://github.com/vllm-project/vllm/pull/3095/files

Co-authored-by: Robert Shaw <114415538+robertgshaw2-neuralmagic@users.noreply.github.com>

[Bug fix][Core] assert num_new_tokens == 1 fails when SamplingParams.n is not 1 and max_tokens is large & Add tests for preemption (#4451)

Co-authored-by: Robert Shaw <114415538+robertgshaw2-neuralmagic@users.noreply.github.com>
Co-authored-by: Robert Shaw <rshaw@neuralmagic.com>

Co-authored-by: Danny Guinther <dguinther@neuralmagic.com>