text-generation-inference
Add support for stop words in TRTLLM
#2678
Merged

Add support for stop words in TRTLLM #2678

Narsil merged 18 commits into main from trtllm-stop-words
mfuntowicz
mfuntowicz mfuntowicz requested a review from Narsil Narsil 1 year ago
mfuntowicz mfuntowicz requested a review from Hugoch Hugoch 1 year ago
mfuntowicz feat(trtllm): rewrite health to not account for current state
7f383bf4
mfuntowicz chore(looper): cleanup a bit more
3af45189
mfuntowicz feat(post_processing): max_new_tokens is const evaluated now
17073267
mfuntowicz chore(ffi):formatting
e711947e
mfuntowicz feat(trtllm): add stop words handling
171a5638
mfuntowicz chore(trtllm): create specific parallelconfig factory and logging ini…
16bb4b67
mfuntowicz chore(trtllm): define a macro for SizeType cast
c90680ed
mfuntowicz chore(trtllm): use GetParallelConfig
703c26ec
mfuntowicz chore(trtllm): minor refactoring
c2bb199f
mfuntowicz chore(trtllm): validate there are enough GPus on the system for the d…
d659cb01
mfuntowicz chore(trtllm): ensure max throughput scheduling policy is selected
cdba16fd
mfuntowicz chore(trtllm): minor fix
b939a0f7
mfuntowicz chore(router): minor refactorings
50a19aa3
mfuntowicz feat(docker): build with-slurm ompi
17573d42
mfuntowicz feat(docker): add python3.10 dev to runtime deps
e4f67f70
mfuntowicz chore(docker): add mpi to ld_library_path
38126233
mfuntowicz chore(docker): install transformers
13a68e22
mfuntowicz feat(trtllm): detect stop_words from generation_config.json
84f14a14
Narsil Narsil force pushed from 9cee00ee to 84f14a14 1 year ago
Narsil Narsil merged ba5fc7d9 into main 1 year ago
Narsil Narsil deleted the trtllm-stop-words branch 1 year ago

Login to write a write a comment.

Login via GitHub

Reviewers
Assignees
No one assigned
Labels
Milestone