Junhao b73b5b0629 Make microbatch optimization (DBO) work with general models (#37926) ...

Signed-off-by: Junhao Li <junhao@ubicloud.com>

2026-03-24 14:40:08 -07:00

..

gpu

[MRV2] Fix for DS v3.2 (#38030)

2026-03-24 14:03:24 -07:00

__init__.py

[V1] Implement vLLM V1 [1/N] (#9289)

2024-10-22 01:24:07 -07:00

block_table.py

Fix Mamba state corruption from referencing stale block table entries (#37728) (#37728) (#37728)

2026-03-24 10:29:59 -07:00

cp_utils.py

[V1][Hybrid] Mamba Prefix Caching with align mode (#30877)

2026-01-23 09:56:48 -08:00

cpu_model_runner.py

[Bugfix] Add replacement of _compute_slot_mapping_kernel on CPU (#37987)

2026-03-24 07:00:20 -07:00

cpu_worker.py

[CPU][UX] Do not crash when tcmalloc/libiomp are not ldpreloaded (#37561)

2026-03-19 16:35:45 +00:00

dp_utils.py

[Log] Reduce duplicate log (#37313)

2026-03-18 10:57:44 -04:00

ec_connector_model_runner_mixin.py

[Misc] Introduce ec_both role EC (encoder cache) connector (#34182)

2026-02-10 18:55:35 +00:00

gpu_input_batch.py

Fix Mamba state corruption from referencing stale block table entries (#37728) (#37728) (#37728)

2026-03-24 10:29:59 -07:00

gpu_model_runner.py

Fix Mamba state corruption from referencing stale block table entries (#37728) (#37728) (#37728)

2026-03-24 10:29:59 -07:00

gpu_ubatch_wrapper.py

Make microbatch optimization (DBO) work with general models (#37926)

2026-03-24 14:40:08 -07:00

gpu_worker.py

elastic_ep: Fix issues with repeated scale up/down cycles (#37131)

2026-03-20 23:13:02 +00:00

kv_connector_model_runner_mixin.py

[Bugfix] Disable cross-layer KV cache for MLA attention backends (#37090)

2026-03-16 19:03:10 +02:00

lora_model_runner_mixin.py

Reduce the kernel overhead when num of active loras is smaller than max loras. Multiple cuda graphs are captured for each num of active-loras. (#32005)

2026-02-02 12:30:06 -05:00

mamba_utils.py

[V0 Deprecation] Refactor kv cache from list to element (#37487)

2026-03-23 20:10:11 -07:00

tpu_input_batch.py

[Cleanup] Remove unused ModelRunner V1 InputBatch.num_tokens field (#30218)

2025-12-18 09:17:00 -08:00

ubatch_utils.py

[6/N][Attention] Move utils to more appropriate locations (#32215)

2026-01-13 05:38:52 -08:00

ubatching.py

[Misc] Replace Optional[X] with X | None syntax (#33332)

2026-01-30 01:56:59 -08:00

utils.py

[V0 Deprecation] Refactor kv cache from list to element (#37487)

2026-03-23 20:10:11 -07:00

worker_base.py

elastic_ep: Fix issues with repeated scale up/down cycles (#37131)

2026-03-20 23:13:02 +00:00

workspace.py

[1/N] Elastic EP Milestone 2 (#34861)

2026-02-28 04:46:42 +00:00

xpu_model_runner.py

[XPU] Enable ModelRunnerV2 on XPU (#36078)

2026-03-05 17:19:18 +00:00

xpu_worker.py

[Hardware] Replace torch.cuda.device_count/current_device/set_device API (#36145)

2026-03-12 07:57:47 -07:00