Lianmin Zheng b94ac87e0c Enable breakable prefill CUDA graph for DP attention (#30898)

2026-07-12 17:10:04 -07:00

..

test_cuda_graph_buffer_registry.py

Enable breakable prefill CUDA graph for DP attention (#30898)

2026-07-12 17:10:04 -07:00

test_forward_metadata_plan_record.py

Replace skip_attn_backend_init with a batch-carried attention plan marker (+ staleness re-plan) (#27193)

2026-06-04 17:13:18 -07:00

test_ngram_token_table.py

[Model] Support LongCat 2.0 FP8 (#30275)

2026-07-07 19:51:12 +08:00

test_pool_configurator.py

[Spec] Add DSpark: confidence-scheduled speculative decoding (#30261)

2026-07-12 17:25:26 -05:00