Giancarlo Delfin d63c8e9444 [BugFix][Spec Decode] Compact shared topk indices buffer after first MTP draft step (#47238)

2026-07-01 21:38:51 -07:00

..

dynamic

[V1][Spec Decode] Add Dynamic SD (#32374)

2026-06-14 00:14:27 -07:00

__init__.py

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

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

custom_class_proposer.py

[Feature] Support custom callable proposer backend for speculative decoding (#39487)

2026-05-13 16:53:01 +00:00

dflash.py

[CPU][Spec Decode] Enable DFlash SD for CPU (#44029)

2026-06-25 15:32:48 +08:00

draft_model.py

[Spec Decode] Move SpecDecodeBaseProposer out of eagle.py (#40732)

2026-04-23 22:28:27 +00:00

eagle.py

[Spec Decode] Move SpecDecodeBaseProposer out of eagle.py (#40732)

2026-04-23 22:28:27 +00:00

extract_hidden_states.py

[EPLB] Mask padding in EPLB load recording (#38128)

2026-06-28 19:43:58 -07:00

gemma4.py

[Bugfix] Fix Gemma4 MTP block_table batch_size mismatch under concurrent load (#43982)

2026-06-03 17:11:10 -07:00

llm_base_proposer.py

[BugFix][Spec Decode] Compact shared topk indices buffer after first MTP draft step (#47238)

2026-07-01 21:38:51 -07:00

medusa.py

[V1][Spec Decode] Add Dynamic SD (#32374)

2026-06-14 00:14:27 -07:00

metadata.py

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

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

metrics.py

[V1][Spec Decode] Add Dynamic SD (#32374)

2026-06-14 00:14:27 -07:00

ngram_proposer_gpu.py

[Platform] Replace torch.cuda.Event with torch.Event (#47140)

2026-06-30 08:39:59 -07:00

ngram_proposer.py

[V1][Spec Decode] Add Dynamic SD (#32374)

2026-06-14 00:14:27 -07:00

step3p5.py

[V1][Spec Decode] Add Dynamic SD (#32374)

2026-06-14 00:14:27 -07:00

suffix_decoding.py

[V1][Spec Decode] Add Dynamic SD (#32374)

2026-06-14 00:14:27 -07:00

utils.py

[CPU] Skip Triton kernel monkey-patches when Triton-CPU is available (#44991)

2026-06-18 18:59:30 +08:00