| .. |
|
__init__.py
|
9d81716bfd
[v0.5.3] Release Candidate (#388)
|
vor 8 Monaten |
|
batch_expansion.py
|
3bb0f07461
chore: rename `task_handler` to `worker` (#985)
|
vor 2 Wochen |
|
draft_model_runner.py
|
5c3b94de45
spec decode: move ops.advane_step to flash attention backend (#1005)
|
vor 2 Wochen |
|
interfaces.py
|
f1d0b77c92
[0.6.0] Release Candidate (#481)
|
vor 4 Monaten |
|
medusa_worker.py
|
3bb0f07461
chore: rename `task_handler` to `worker` (#985)
|
vor 2 Wochen |
|
metrics.py
|
2f61644f6e
SPMD optimizations (#824)
|
vor 1 Monat |
|
mlp_speculator_worker.py
|
0dfa6b60ec
core: support logprobs with multi-step scheduling (#963)
|
vor 2 Wochen |
|
multi_step_worker.py
|
3bb0f07461
chore: rename `task_handler` to `worker` (#985)
|
vor 2 Wochen |
|
ngram_worker.py
|
0dfa6b60ec
core: support logprobs with multi-step scheduling (#963)
|
vor 2 Wochen |
|
proposer_worker_base.py
|
3bb0f07461
chore: rename `task_handler` to `worker` (#985)
|
vor 2 Wochen |
|
smaller_tp_proposer_worker.py
|
0dfa6b60ec
core: support logprobs with multi-step scheduling (#963)
|
vor 2 Wochen |
|
spec_decode_worker.py
|
3bb0f07461
chore: rename `task_handler` to `worker` (#985)
|
vor 2 Wochen |
|
target_model_runner.py
|
3bb0f07461
chore: rename `task_handler` to `worker` (#985)
|
vor 2 Wochen |
|
top1_proposer.py
|
0dfa6b60ec
core: support logprobs with multi-step scheduling (#963)
|
vor 2 Wochen |
|
util.py
|
0dfa6b60ec
core: support logprobs with multi-step scheduling (#963)
|
vor 2 Wochen |
AlpinDale
5c3b94de45
spec decode: move ops.advane_step to flash attention backend (#1005)