vllmellm e9717801bd [Bugfix][ROCm] Fix Unsupported attention metadata type for speculative decoding in eagle.py (#31714) ...

Signed-off-by: vllmellm <vllm.ellm@embeddedllm.com>

2026-01-06 07:53:22 +00:00

..

__init__.py

[V1][BugFix] Add __init__.py to v1/spec_decode/ (#13359)

2025-02-16 09:39:08 -08:00

eagle.py

[Bugfix][ROCm] Fix Unsupported attention metadata type for speculative decoding in eagle.py (#31714)

2026-01-06 07:53:22 +00:00

medusa.py

[V1][Spec Decode] Optimize Medusa proposer to avoid GPU-CPU sync (#29723)

2025-12-10 00:15:01 +00:00

metadata.py

[V1][spec decode] return logprobs for spec decoding (#26060)

2025-10-22 22:59:59 -07:00

metrics.py

[mypy] Enable type checking for more directories (#29674)

2025-11-28 08:39:27 -08:00

ngram_proposer.py

[BugFix] Fix index error in ngram_proposer (#29779)

2025-12-02 04:48:11 +00:00

suffix_decoding.py

Revert "[Redo] #26368 (#28771)" (#29121)

2025-11-20 21:27:45 -08:00

utils.py

[SpecDecode] Simplified alternative padded-speculation acceptance rate fix (#29845)

2025-12-22 13:06:10 -08:00