Alex NailsandClaude Opus 5 bfa4e4a57b [Nemotron] Hoist mamba track-mask host syncs out of the per-layer prefill path (#32589) ...

Co-authored-by: Claude Opus 5 (1M context) <noreply@anthropic.com>

2026-08-03 23:58:28 -07:00

..

model_runner_components

fix(server): capture legal multi-request prefill CUDA graph batches (#30206)

2026-08-03 16:56:59 -07:00

runner

[Nemotron] Hoist mamba track-mask host syncs out of the per-layer prefill path (#32589)

2026-08-03 23:58:28 -07:00

runner_backend

[BCG][3/N] Enable bcg on dsa & deepep a2a backend (#31987)

2026-07-31 16:45:18 -07:00

runner_backend_utils

[BCG][3/N] Enable bcg on dsa & deepep a2a backend (#31987)

2026-07-31 16:45:18 -07:00

runner_utils

[Spec] Rename num_tokens_per_bs to num_tokens_per_req (#30977)

2026-07-13 13:47:53 -05:00

cpu_graph_runner.py

[Feature] Support return_hidden_states="last" (#30177)

2026-08-02 15:09:33 +08:00

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

cuda_graph_config.py

fix(server): capture legal multi-request prefill CUDA graph batches (#30206)

2026-08-03 16:56:59 -07:00

forward_batch_deepseek_mha_mixin.py

[Kernel] Introduce sglang.kernels namespace and migrate scattered triton_ops kernels (RFC #29630, Phase 2) (#30044)

2026-07-10 21:41:08 +08:00

forward_batch_info.py

[Feature] Support return_hidden_states="last" (#30177)

2026-08-02 15:09:33 +08:00

forward_context.py

[Refactor] Cuda Graph Runner/Backend Refactor (#23906)

2026-06-09 21:36:57 -07:00

graph_shared_output.py

Share one logits output buffer across prefill/decode/draft cuda-graph runners (#29779)

2026-07-01 23:30:56 -07:00

hook_manager.py

Rename: --hooks to --forward-hooks (#13994)

2025-11-26 22:26:28 +08:00

input_buffers.py

[refactor] Retire DecodeInputBuffers / PrefillInputBuffers in favor of CudaGraphBufferRegistry (#27192)

2026-06-03 20:52:56 -07:00

mindspore_runner.py

[NPU]mindspore model support moe (#15363)

2026-02-02 17:52:49 +08:00

model_runner.py

[CP] Support breakable CUDA graphs for zigzag strategy (#33136)

2026-08-02 23:11:57 -07:00

pool_configurator.py

[Kimi] Support DCP + DSpark (ported from kimi-k3 branch) (#32828)

2026-07-31 17:39:00 -07:00