AlpinDale 16dff9babc chore: enable bonus token in spec decoding for KV cache based models 6 mesiacov pred
..
__init__.py 9d81716bfd [v0.5.3] Release Candidate (#388) 10 mesiacov pred
batch_expansion.py af43576da0 feat: add MLPSpeculator speculative decoding support (#572) 7 mesiacov pred
draft_model_runner.py 99680b2d23 feat: soft prompts (#589) 6 mesiacov pred
interfaces.py 16dff9babc chore: enable bonus token in spec decoding for KV cache based models 6 mesiacov pred
medusa_worker.py 16dff9babc chore: enable bonus token in spec decoding for KV cache based models 6 mesiacov pred
metrics.py 7253e9052d feat: integrate typical acceptance sampling for spec decoding 7 mesiacov pred
mlp_speculator_worker.py 16dff9babc chore: enable bonus token in spec decoding for KV cache based models 6 mesiacov pred
multi_step_worker.py 16dff9babc chore: enable bonus token in spec decoding for KV cache based models 6 mesiacov pred
ngram_worker.py 16dff9babc chore: enable bonus token in spec decoding for KV cache based models 6 mesiacov pred
proposer_worker_base.py 16dff9babc chore: enable bonus token in spec decoding for KV cache based models 6 mesiacov pred
smaller_tp_proposer_worker.py 16dff9babc chore: enable bonus token in spec decoding for KV cache based models 6 mesiacov pred
spec_decode_worker.py 16dff9babc chore: enable bonus token in spec decoding for KV cache based models 6 mesiacov pred
top1_proposer.py 16dff9babc chore: enable bonus token in spec decoding for KV cache based models 6 mesiacov pred
util.py af43576da0 feat: add MLPSpeculator speculative decoding support (#572) 7 mesiacov pred