kk f1384f5293 Integration mori backend for EP a2a data communication (#17012) ...

Co-authored-by: Duyi-Wang <duyi.wang@amd.com>
Co-authored-by: billishyahao <bill.he@amd.com>
Co-authored-by: HaiShaw <hixiao@gmail.com>

2026-01-28 19:07:34 -08:00

..

async_dynamic_batch_tokenizer.py

[Performance] Dynamic Batch Tokenizer (#9382)

2025-09-14 01:56:04 +08:00

async_mm_data_processor.py

[VLM] Optimize async mm data process mechanism (#12066)

2025-11-01 01:24:53 +08:00

cache_controller.py

[HiCache][HA 1/N] Support HiCache storage runtime attach/detach (#15892)

2026-01-26 19:33:19 -08:00

configure_logging.py

Improve streaming, log_level, memory report, weight loading, and benchmark script (#7632)

2025-06-29 23:16:19 -07:00

data_parallel_controller.py

[3/N] Achieve fault tolerance at the DP level (#11657)

2026-01-20 18:47:08 +08:00

detokenizer_manager.py

Special logic for healthcheck (#17734)

2026-01-26 10:26:40 -08:00

disagg_service.py

refactor(codespell): corrects typos covered up by whitelist (#17601)

2026-01-22 11:45:47 -08:00

io_struct.py

[HiCache][HA 1/N] Support HiCache storage runtime attach/detach (#15892)

2026-01-26 19:33:19 -08:00

mm_utils.py

use shared memory for multimodal feature transport between Tokenizer and Scheduler (#16402)

2026-01-27 11:01:08 -08:00

multi_tokenizer_mixin.py

[Auto Sync] Update detokenizer_manager.py, io_struct.py, mu... (20260120) (#17442)

2026-01-21 14:48:32 -08:00

multimodal_processor.py

Skip mm feature pool init to avoid EPD OOM (#16388)

2026-01-23 10:53:45 +08:00

overlap_utils.py

feat(SpecEagleV2): add standalone_worker_v2 (#12625)

2025-12-30 17:55:04 +08:00

prefill_delayer.py

Support token low usage watermark in prefill delayer (#16814)

2026-01-09 23:07:18 +08:00

request_metrics_exporter.py

[Auto Sync] Update request_metrics_exporter.py (20251230) (#16183)

2025-12-30 11:09:41 -08:00

schedule_batch.py

Introduce global alloc_len_per_decode & clean check decode memory (#15115)

2026-01-26 10:26:20 -08:00

schedule_policy.py

refactor(codespell): corrects typos covered up by whitelist (#17601)

2026-01-22 11:45:47 -08:00

scheduler_dp_attn_mixin.py

Add flag that enables NCCL mlp sync batch for overlap scheduler (#17288)

2026-01-20 23:06:55 +08:00

scheduler_input_blocker.py

Tiny move files to utils folder (#11166)

2025-10-03 22:40:06 +08:00

scheduler_metrics_mixin.py

Integration mori backend for EP a2a data communication (#17012)

2026-01-28 19:07:34 -08:00

scheduler_output_processor_mixin.py

[Auto Sync] Update detokenizer_manager.py, io_struct.py, mu... (20260120) (#17442)

2026-01-21 14:48:32 -08:00

scheduler_pp_mixin.py

[PD] Fix PP dynamic chunking with DP attention (#17339)

2026-01-20 11:27:13 +08:00

scheduler_profiler_mixin.py

[NPU] torch_npu profiler tensorboard path type fix (#17545)

2026-01-24 22:55:49 +08:00

scheduler_recv_skipper.py

[Feat]Add scheduler recv skipper weights to environment configuration (#13855)

2025-11-27 18:16:11 +08:00

scheduler_runtime_checker_mixin.py

Refactor prefix cache type checking (#17028)

2026-01-15 11:28:13 +08:00

scheduler_update_weights_mixin.py

refactor: add type hints to scheduler mixins (#15913)

2025-12-26 16:50:07 -08:00

scheduler.py

use shared memory for multimodal feature transport between Tokenizer and Scheduler (#16402)

2026-01-27 11:01:08 -08:00

session_controller.py

refactor(codespell): corrects typos covered up by whitelist (#17601)

2026-01-22 11:45:47 -08:00

template_manager.py

fix: Handle multiple named chat templates in HuggingFace tokenizers (#17236)

2026-01-18 17:20:04 +08:00

tokenizer_communicator_mixin.py

[HiCache][HA 1/N] Support HiCache storage runtime attach/detach (#15892)

2026-01-26 19:33:19 -08:00

tokenizer_manager_multiitem_mixin.py

feat: support qwen3(-VL) rerank scoring&chat template (#16403)

2026-01-15 00:45:46 +08:00

tokenizer_manager.py

use shared memory for multimodal feature transport between Tokenizer and Scheduler (#16402)

2026-01-27 11:01:08 -08:00

tp_worker.py

add documentation example for LoRA overlap loading and cleanup unused function (#17464)

2026-01-24 15:33:16 +08:00

utils.py

Introduce global alloc_len_per_decode & clean check decode memory (#15115)

2026-01-26 10:26:20 -08:00