transformers
[serving] Fix continuous batching JSON response serialization
#45057
Merged

Commits
  • Fix continuous batching JSON response serialization
    NathanHB committed 14 days ago
  • add example script eval-job
    NathanHB committed 13 days ago
  • fix script
    NathanHB committed 13 days ago
  • Add test for continuous batching non-streaming JSON response
    NathanHB committed 13 days ago
  • fix ci
    NathanHB committed 13 days ago
  • Update eval script to use official transformers repo main branch
    NathanHB committed 13 days ago
  • add kernels and flash attn 2
    NathanHB committed 10 days ago
  • Add continuous batching configuration CLI arguments to serve command
    NathanHB committed 10 days ago
  • Add thread lock for manager creation to avoid double manager
    remi-or committed 9 days ago
  • change transformers dep
    NathanHB committed 9 days ago
Loading