3 people f8f2870a84 Profiling Enhancements [1/3]: cuda graph profile traces (#24370) ...

Co-authored-by: Basit <mohbasit@ctr2-alola-ctrl-01.amd.com>
Co-authored-by: HAI <hixiao@gmail.com>

2026-08-06 03:19:35 -07:00

..

model_runner_components

[CUDA Graph] Allow custom decode graph runners (#33553)

2026-08-04 12:48:56 -07:00

runner

Profiling Enhancements [1/3]: cuda graph profile traces (#24370)

2026-08-06 03:19:35 -07:00

runner_backend

Profiling Enhancements [1/3]: cuda graph profile traces (#24370)

2026-08-06 03:19:35 -07:00

test_chunked_prefix_cache_gate.py

config: the draft runner carries its own attention backend

2026-08-05 19:32:24 -07:00

test_cuda_graph_buffer_registry.py

Fix pad-row top-k masking with custom_routing_function under DP attention (#31838)

2026-07-21 11:49:17 -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_hisparse_pool_configurator.py

fix(hisparse): correct DSA KV memory budget (#31992)

2026-07-27 11:19:32 +08:00

test_mlp_sync_pad_unpad.py

Multiple flexibility fixes for DP attention (#33537)

2026-08-04 15:40:40 -07:00

test_pool_configurator.py

test: recover the config-namespace-migration deferrals (#33171)

2026-08-01 08:59:27 -07:00

test_prefill_cuda_graph_runner_helpers.py

Enable multimodal prefill BCG for VL and audio models (#30872)

2026-07-29 06:47:40 +08:00

test_prefill_cuda_graph_runner.py

config: resolve the draft worker's config per runner, not on a copy

2026-08-05 19:31:23 -07:00