Skip to content

Add DFlash speculative decoding support to NeMo SpeechLM vLLM plugin - #16129

Closed
Slyne wants to merge 20 commits into
NVIDIA-NeMo:mainfrom
Slyne:slyne/dflash-inference
Closed

Add DFlash speculative decoding support to NeMo SpeechLM vLLM plugin#16129
Slyne wants to merge 20 commits into
NVIDIA-NeMo:mainfrom
Slyne:slyne/dflash-inference

feat(speechlm2): route Automodel DFlash2 through vLLM

e095c3c
Select commit
Loading
Failed to load commit list.
GitHub Advanced Security / CodeQL succeeded Sep 4, 2026 in 3s

1 new alert

New alerts in code changed by this pull request

  • 1 note

See annotations below for details.

View all branch alerts.

Annotations

Check notice on line 160 in nemo/collections/speechlm2/vllm/salm/audio.py

See this annotation in the file changed.

Code scanning / CodeQL

Empty except Note

'except' clause does nothing but pass and there is no explanatory comment.