Chendi.Xue 0a71900bc9 Remove hard-dependencies of Speculative decode to CUDA workers (#10587) ...

Signed-off-by: Chendi Xue <chendi.xue@intel.com>

2024-11-26 17:57:11 -08:00

..

e2e

[Bugfix] Fix for Spec model TP + Chunked Prefill (#10232)

2024-11-26 09:11:16 -08:00

__init__.py

[Speculative decoding 3/9] Worker which speculates, scores, and applies rejection sampling (#3103)

2024-03-08 23:32:46 -08:00

test_batch_expansion.py

[BugFix] Fix use of per-request seed with pipeline parallel (#6698)

2024-07-30 10:40:08 -07:00

test_dynamic_spec_decode.py

[Spec Decode] Disable Log Prob serialization to CPU for spec decoding for both draft and target models. (#6485)

2024-07-20 23:58:58 -07:00

test_metrics.py

[Bugfix] [SpecDecode] AsyncMetricsCollector: update time since last collection (#6578)

2024-07-19 14:01:03 -07:00

test_multi_step_worker.py

[Bugfix][SpecDecode] kv corruption with bonus tokens in spec decode (#9730)

2024-11-06 01:45:45 +00:00

test_ngram_worker.py

[Feature] [Spec decode]: Combine chunked prefill with speculative decoding (#9291)

2024-11-07 08:15:14 -08:00

test_scorer.py

[Feature] [Spec decode]: Combine chunked prefill with speculative decoding (#9291)

2024-11-07 08:15:14 -08:00

test_spec_decode_worker.py

Remove hard-dependencies of Speculative decode to CUDA workers (#10587)

2024-11-26 17:57:11 -08:00

test_utils.py

[Bugfix] Unify rank computation across regular decoding and speculative decoding (#7899)

2024-08-28 22:18:13 -07:00

utils.py

[Feature] [Spec decode]: Combine chunked prefill with speculative decoding (#9291)

2024-11-07 08:15:14 -08:00