Yuanle Liu 326b9755aa [BugFix][MTP] Skip empty_input_forward during dummy run (#6653) ...

When `is_dummy_run=True`, calling `empty_input_forward` can cause
unexpected behavior. Add `and not is_dummy_run` guard for both
`_propose_cuda` and `_propose_xpu` paths.

Co-authored-by: Claude Sonnet 4.6 <noreply@anthropic.com>

2026-03-04 23:51:56 -08:00

..

__init__.py

[Speculative Decoding] Support suffix decoding (#6403)

2026-02-26 11:42:05 +08:00

base.py

[Speculative Decoding] Support suffix decoding (#6403)

2026-02-26 11:42:05 +08:00

mtp.py

[BugFix][MTP] Skip empty_input_forward during dummy run (#6653)

2026-03-04 23:51:56 -08:00

ngram.py

[Feature] GPU Memory Optimization and Retirement of V0 Scheduler (#6407)

2026-02-28 15:07:43 +08:00

suffix.py

[Speculative Decoding] Support suffix decoding (#6403)

2026-02-26 11:42:05 +08:00