[#9760][fix] Use RequestError for validation errors to prevent engine shutdown#9761

Open

tzulingk wants to merge 4 commits intoNVIDIA:mainfrom

tzulingk:tzulingk/request_error

Commits on Dec 6, 2025

Fix: Change max_num_tokens validation error from ValueError to RequestError to prevent engine shutdown
tzulingk
committed

Commits on Dec 9, 2025

Fix thread leak in test_max_num_token_check by using RequestError and adding proper cleanup
tzulingk
committed

Commits on Dec 15, 2025

Fix tests to expect RequestError for max_num_tokens validation in PyTorch backend
tzulingk
committed

Commits on Dec 16, 2025

Merge branch 'main' into tzulingk/request_error
karljang
authored