3 people 957d512e96 [Bugfix][Frontend] Preserve streaming chunk boundaries in RequestOutputCollector (#58941) ...

Signed-off-by: David Cheung <d7cheung@gmail.com>
Co-authored-by: Benedikt Falk <bfalk@nvidia.com>
Co-authored-by: Nick Hill <nickhill123@gmail.com>

2026-10-15 16:38:28 +00:00

..

attention

[Perf] Fuse the mamba align block-table gather into one Triton kernel, 7 kernels -> 1, 6x speedup on all shapes for this kernel (#58737)

2026-10-15 11:02:53 -05:00

core

[Core] Apply prefill_schedule_interval outside data parallelism (#54627)

2026-10-15 11:03:13 -05:00

engine

[Bugfix][Frontend] Preserve streaming chunk boundaries in RequestOutputCollector (#58941)

2026-10-15 16:38:28 +00:00

executor

[Core] Remove the legacy Ray compiled-graph executor (#61712)

2026-10-14 18:13:27 -07:00

fault_tolerance

[Bugfix][FT] Pass stateless process-group timeouts explicitly (#58611)

2026-09-29 03:14:02 +00:00

hisparse

[HiSparse] Prefault the shared host pool in parallel (#60869)

2026-10-14 10:28:26 -04:00

kv_hints

[Feature] Add first-class KV hints request envelope for programmatic KV management (#53423)

2026-09-21 09:56:11 -07:00

kv_offload

[Bugfix][KV Offload] Always attempt NIXL agent removal during peer teardown (#60279)

2026-10-15 14:07:23 +03:00

metrics

[Perf] Read VLLM_COMPUTE_NANS_IN_LOGITS once per IterationStats (#61710)

2026-10-15 18:35:54 +08:00

pool

[Bugfix][Pooling] Handle multimodal cache misses without crashing (#58975)

2026-10-02 10:14:33 -04:00

sample

[Perf] Improve _topk_topp_kernel() for masked logits (#61213)

2026-10-12 16:01:11 -07:00

simple_kv_offload

[Bugfix][KV Offload] Honor speculative cacheability in SimpleCPU (#60071)

2026-10-06 14:25:08 +02:00

spec_decode

[Chore] Use upstream Transformers DeepseekV4Config (#61392)

2026-10-15 01:08:08 -07:00

structured_output

[Bugfix] Mask special added tokens in xgrammar JSON, regex and EBNF grammars (#60749)

2026-10-13 14:38:23 +03:00

watermarking

[watermark] compatibility validation (#56801)

2026-10-05 14:37:13 -04:00

worker

[Bugfix][CPU] Report the real cause when the worker outgrows its memory reservation (#59472)

2026-10-15 21:52:05 +08:00

__init__.py

…

cudagraph_dispatcher.py

[Attention] Remove cascade attention (#61126)

2026-10-12 10:41:07 -07:00

kv_cache_interface.py

[KV Cache] GLM-5.3-Flash: use the generic packed KV layout; CircularBufferSpec tail with speculative ring slots (#57169)

2026-10-09 15:48:37 -04:00

kv_cache_layout.py

[CI] Ignore ruff D209, rejoin the docstrings it split, and silence incompatible-rule warnings (#57212)

2026-09-16 10:57:24 -07:00

kv_cache_spec_registry.py

Add pydocstyle to the ruff rules (#52136)

2026-09-16 16:44:29 +00:00

outputs.py

[Feature][Spec Decode] Support sampling mask replay for MRV2 MTP (#59359)

2026-10-01 13:51:39 +08:00

request.py

[Feature] Add first-class KV hints request envelope for programmatic KV management (#53423)

2026-09-21 09:56:11 -07:00

serial_utils.py

[Feature] Per-row candidate IDs for prefill token scoring (M2 of #56860) (#56984)

2026-10-03 14:17:09 +00:00

utils.py

[Frontend] Accept each connection on the least loaded API worker (#60723)

2026-10-09 04:40:13 +00:00