Augusto Yao ea8f4e9f3f [feature] implement dcp for deepseek_v2 (#14194)

2026-06-25 15:15:04 -07:00

..

runner

[feature] implement dcp for deepseek_v2 (#14194)

2026-06-25 15:15:04 -07:00

runner_backend

[Refactor] Share CUDA graph memory pool across prefill and decode (#28973)

2026-06-24 10:18:20 -07:00

runner_backend_utils

Clean up CUDA graph capture logs (#28937)

2026-06-22 06:15:26 -07:00

runner_utils

[Refactor] Share CUDA graph memory pool across prefill and decode (#28973)

2026-06-24 10:18:20 -07:00

triton_ops

[codex] Centralize Triton utility kernels (#26000)

2026-06-02 16:47:45 +08:00

cpu_graph_runner.py

refactor(runner): rename runner replay/load/can_run for the shared surface (#28384)

2026-06-19 01:45:04 -07:00

cuda_graph_buffer_registry.py

refactor(runner): add EagerRunner, own the eager path, polymorphic dispatch (#28386)

2026-06-19 13:05:43 -07:00

cuda_graph_config.py

[Tiny] Cuda Graph Refactor Code Style Follow up (#28107)

2026-06-12 18:38:13 -07:00

forward_batch_deepseek_mha_mixin.py

[feature] implement dcp for deepseek_v2 (#14194)

2026-06-25 15:15:04 -07:00

forward_batch_info.py

[feature] implement dcp for deepseek_v2 (#14194)

2026-06-25 15:15:04 -07:00

forward_context.py

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

2026-06-09 21:36:57 -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_kv_cache_mixin.py

[feature] implement dcp for deepseek_v2 (#14194)

2026-06-25 15:15:04 -07:00

model_runner.py

[fix] Skip routed expert capture for draft model under spec v2 (#26980)

2026-06-24 16:56:07 -07:00

pool_configurator.py

[Spec] Dissolve EagleDraftInputV2Mixin so spec-info dataclasses hold data only (#29220)

2026-06-24 18:08:30 -07:00