Lianmin Zheng 4319978c73 Fix data parallel perf regression (#6183)

2025-05-10 19:18:35 -07:00

..

multimodal_processors

[Fix] Incorrect Memory Allocation on CUDA:0 by Non-Zero CUDA Processes in TP/DP (#5745)

2025-05-08 17:52:26 -07:00

cache_controller.py

fix bug that gpu0 occupies more memory when hicache is turned on (#5778)

2025-05-09 15:36:08 -07:00

configure_logging.py

Support penalty in overlap mode; return logprob with chunked prefill; improve benchmark scripts (#3988)

2025-03-03 00:12:04 -08:00

data_parallel_controller.py

Fix data parallel perf regression (#6183)

2025-05-10 19:18:35 -07:00

detokenizer_manager.py

Clean up imports (#5467)

2025-04-16 15:26:49 -07:00

expert_distribution.py

Support compressed tensors fp8w8a8 (#4743)

2025-03-26 13:21:25 -07:00

io_struct.py

[PD] Add control to slow down a server (#5572)

2025-05-08 01:03:08 -07:00

mm_utils.py

vlm: enable radix cache for qwen-vl models (#5349)

2025-04-23 20:35:05 -07:00

multimodal_processor.py

Clean up imports (#5467)

2025-04-16 15:26:49 -07:00

schedule_batch.py

[Fix] Incorrect Memory Allocation on CUDA:0 by Non-Zero CUDA Processes in TP/DP (#5745)

2025-05-08 17:52:26 -07:00

schedule_policy.py

Fix problem of large page size with chunked prefill (#6046)

2025-05-06 15:19:47 +08:00

scheduler_output_processor_mixin.py

[PP] Add pipeline parallelism (#5724)

2025-04-30 18:18:07 -07:00

scheduler.py

[PD] Add control to slow down a server (#5572)

2025-05-08 01:03:08 -07:00

session_controller.py

model: Minicpmo (#3023)

2025-03-24 20:08:40 -07:00

tokenizer_manager.py

[PD] Add control to slow down a server (#5572)

2025-05-08 01:03:08 -07:00

tp_worker_overlap_thread.py

[PP] Add pipeline parallelism (#5724)

2025-04-30 18:18:07 -07:00

tp_worker.py

Tiny refactor ModelConfig.from_server_args (#5219)

2025-05-08 01:02:43 -07:00

utils.py

refactor: multimodal data (#4754)

2025-03-31 09:57:51 -07:00