R0CKSTARandMick db97f193b7 [diffusion][llm] macOS support (#19549) ...

Signed-off-by: Xiaodong Ye <yeahdongcn@gmail.com>
Co-authored-by: Mick <mickjagger19@icloud.com>

2026-03-10 13:11:07 -07:00

..

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

Fix mooncake store write/read bandwidth logs (#18294)

2026-03-09 19:54:37 -07:00

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

[PD-Disagg] Fully support external DP dispatch w/ PD-disaggregation mode. (#19268)

2026-02-24 19:58:01 -08:00

detokenizer_manager.py

[PD-Disagg] Fully support external DP dispatch w/ PD-disaggregation mode. (#19268)

2026-02-24 19:58:01 -08:00

disagg_service.py

[PD] Cleanup BootstrapServer init and ready check (#19551)

2026-02-28 16:41:42 +08:00

io_struct.py

[feat] Enhance lora_update_weight_from_tensor for RL training (#19314)

2026-03-04 18:10:42 -08:00

mm_utils.py

Fix EPD OOM by offloading precomputed_embeddings during chunked prefill (#16503)

2026-03-08 20:10:40 -07:00

multi_tokenizer_mixin.py

[PD-Disagg] Fully support external DP dispatch w/ PD-disaggregation mode. (#19268)

2026-02-24 19:58:01 -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

[spec v2]Fix torch gc of future indices (#18958)

2026-02-19 11:38:25 -08:00

prefill_delayer.py

[OPT]Skip the first delayer to maximize the BS of the decoding. (#19836)

2026-03-06 08:53:19 +08:00

schedule_batch.py

Enable piecewise-cuda-graph when logprob_start_len = -1 (#19453)

2026-03-10 12:50:57 -07:00

schedule_policy.py

[NPU] bugs fix for Deepseek models (#19544)

2026-02-28 17:26:15 +08:00

scheduler_dp_attn_mixin.py

Optimization: Reduce the number of D2H operations (#19424)

2026-03-04 16:32:42 +08:00

scheduler_input_blocker.py

Tiny move files to utils folder (#11166)

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

scheduler_output_processor_mixin.py

[Performance] Decode Offload improves the long texts performance 100% through dynamic block offload. (#17216)

2026-03-08 17:16:53 +08:00

scheduler_pp_mixin.py

Revert "[Scheduler] Decouple maybe_send_health_check_signal from process_batch_result" (#20259)

2026-03-10 01:58:48 -07:00

scheduler_profiler_mixin.py

Add XPU profiler activity support in benchmark code (#12981)

2026-03-04 23:22:56 -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

Fix streaming session with paged KV cache (SWA/MLA) (#20070)

2026-03-08 03:00:32 -07:00

scheduler_update_weights_mixin.py

[Fix] Add --disable-draft-model-update to control draft model updates(especially in RL) (#15726)

2026-02-28 12:09:55 -08:00

scheduler.py

[diffusion][llm] macOS support (#19549)

2026-03-10 13:11:07 -07:00

session_controller.py

[Session] Extract SessionController and clean up session logic in Scheduler (#19547)

2026-02-28 19:47:44 -08:00

template_manager.py

fix: fix the bug blocking completion template application (#17010)

2026-02-24 18:50:08 +08:00

tokenizer_communicator_mixin.py

feat: TTL-based prefix pinning with refresh-on-hit for HiRadixCache (#18941)

2026-03-01 23:27:22 -08:00

tokenizer_manager_multiitem_mixin.py

[Score API][18132] return token usage in Score API response (#18381)

2026-03-03 13:45:35 -08:00

tokenizer_manager.py

[Bug] Fix missing TTFT histogram for single-batch requests (#20122)

2026-03-08 15:18:51 -07:00

tp_worker.py

[Core] Replace server_args mutation hack with explicit MemoryPoolConfig for draft worker init (#20183)

2026-03-09 11:45:54 -07:00

utils.py

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

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