Aurick QiaoandYuwei An 26fd7fdaa2 [Feature] Support PP in full prefill CUDA graphs (#35451) ...

Co-authored-by: Yuwei An <ayw.sirius19@gmail.com>

2026-08-27 17:32:00 -07:00

..

__init__.py

Introduce WeightUpdater and WeightExporter components (#31148)

2026-07-14 15:53:33 +08:00

attention_backend_setup.py

config: the alias form of the runner-side instance read

2026-08-15 00:39:03 -07:00

cuda_graph_setup.py

[Feature] Support PP in full prefill CUDA graphs (#35451)

2026-08-27 17:32:00 -07:00

kv_pool_runtime.py

[Feature] Add MiniCPM-SALA support (#30360)

2026-08-24 02:25:16 -07:00

layer_setup.py

Fix broken Nemotron DP attention (#33123)

2026-08-05 14:27:05 -07:00

load_model_utils.py

config: the record is not an object that gets passed around (#36622)

2026-08-27 12:57:10 -07:00

misc_utils.py

config: decisions keyed on the attention backend read the configured pair

2026-08-15 00:38:00 -07:00

moe_ep_setup.py

config: the record is not an object that gets passed around (#36622)

2026-08-27 12:57:10 -07:00

ngram_embedding_manager.py

config: stop handing the record to code that does not read it (#36252)

2026-08-26 05:02:17 -07:00

remote_instance_weight_transporter.py

config: a parallel leaf with no live counterpart is read bare (#36620)

2026-08-27 12:56:11 -07:00

spec_aux_hidden_state.py

config: a parallel size has one spelling; a patched scope declares its own (#36621)

2026-08-27 12:56:42 -07:00

startup_weight_load.py

config: a parallel size has one spelling; a patched scope declares its own (#36621)

2026-08-27 12:56:42 -07:00

weight_exporter.py

Introduce WeightUpdater and WeightExporter components (#31148)

2026-07-14 15:53:33 +08:00

weight_updater.py

[FEAT] Support fast engine recovery through weight cache (#27139)

2026-07-25 14:31:21 +08:00