AlpinDale ae04f57ec1 feat: Pipeline Parallel support (#581) vor 7 Monaten
..
__init__.py 9d81716bfd [v0.5.3] Release Candidate (#388) vor 10 Monaten
batch_expansion.py af43576da0 feat: add MLPSpeculator speculative decoding support (#572) vor 7 Monaten
draft_model_runner.py ae04f57ec1 feat: Pipeline Parallel support (#581) vor 7 Monaten
interfaces.py af43576da0 feat: add MLPSpeculator speculative decoding support (#572) vor 7 Monaten
metrics.py 7253e9052d feat: integrate typical acceptance sampling for spec decoding vor 7 Monaten
mlp_speculator_worker.py 405bb74612 Control plane comms refactor (#573) vor 7 Monaten
multi_step_worker.py cdff8e89f9 feat: introduce `DraftModelRunner` vor 7 Monaten
ngram_worker.py e0886ee929 feat: add `ProposerWorkerBase` abstract class vor 7 Monaten
proposer_worker_base.py abbb730607 feat: support draft model on different tensor parallel size vor 7 Monaten
smaller_tp_proposer_worker.py b6ff0623a6 chore: clean up branding vor 7 Monaten
spec_decode_worker.py dd378ea063 feat: MLPSpeculator with tensor parallel vor 7 Monaten
top1_proposer.py af43576da0 feat: add MLPSpeculator speculative decoding support (#572) vor 7 Monaten
util.py af43576da0 feat: add MLPSpeculator speculative decoding support (#572) vor 7 Monaten