Luke AlonsoandLucas Wilkinson 0d4ad47981 [Kernel] Add B12X causal paged attention backend (#52017) ...

Signed-off-by: Luke Alonso <lalonso@gmail.com>
Co-authored-by: Lucas Wilkinson <LucasWilkinson@users.noreply.github.com>

2026-09-01 10:42:02 -04:00

..

attention

[Kernel] Add B12X causal paged attention backend (#52017)

2026-09-01 10:42:02 -04:00

core

[Feature][Spec] Support disabling trailing prefix-cache block dropping (#53388)

2026-09-01 19:12:36 +08:00

cudagraph

[Bugfix][Spec Decode] Capture the widest uniform decode batch by default (#50488)

2026-08-28 22:38:13 -07:00

determinism

[Test][VLM] Add batch-invariance tests for Qwen3-VL (#53531)

2026-08-30 09:49:41 -04:00

distributed

[Bugfix][DP] Synchronize the device on pause completion (#52914)

2026-08-26 09:50:33 -07:00

e2e

[Mamba] Add FlashInfer ReplaySSM backend (#52506)

2026-09-01 10:26:43 -04:00

ec_connector

[Bugfix][EC Connector] Fall back when MADV_POPULATE_WRITE is unsupported (#53190)

2026-09-01 03:17:45 +00:00

engine

[Bugfix][Test] Fix off-by-one error in sampled token rank causing flaky logprobs test (#53976)

2026-08-31 18:49:43 +00:00

executor

[Bugfix] Release worker RPC payload before next dequeue (#51979)

2026-08-24 07:27:36 -07:00

fault_tolerance

[Feature] Add fault tolerance framework (simplified) for DP+EP external LB deployments (#44428)

2026-07-25 11:49:09 -04:00

kv_connector

[DecodeBenchConnector] Fix HMA cache-group mapping (#54647)

2026-09-01 05:52:50 +00:00

kv_offload

[Bugfix][KV Offload] Unlink /dev/shm region after all workers map it (barrier variant of #51317) (#52596)

2026-08-31 15:53:00 +03:00

logits_processors

[Structured Output] Let terminal grammars stop under min_tokens (II) (#54218)

2026-08-31 20:39:27 +04:00

metrics

[Refactor] Move iteration logging to the frontend (#46647)

2026-07-15 17:59:05 -07:00

sample

[Model Runner V2] Reserve CUDA graph memory (#53306)

2026-08-23 23:11:19 -07:00

shutdown

[MyPy][1/N] Fix mypy errors in some tests/ directories and enforce follow-imports=silent (#49570)

2026-07-30 05:02:51 -07:00

simple_kv_offload

[Core][KV Connector] Start async KV loads after the forward launch when no sync loads are scheduled (#53333)

2026-08-28 21:58:00 +00:00

spec_decode

[Bugfix][Model Runner V2][Spec Decode] Decouple the draft's gumbel noise stream from the target's (#54282)

2026-08-29 09:58:29 -07:00

streaming_input

[Spec Decode] DSpark confidence-scheduled verification (#47808)

2026-08-12 12:27:48 -07:00

structured_output

[Bugfix] Support MistralCommonBackend tokenizers in structured output (#52720)

2026-08-20 19:08:50 +01:00

tracing

[CI] Fix segfault in tracing test (#47299)

2026-07-01 14:00:37 +00:00

worker

[Mamba] Add FlashInfer ReplaySSM backend (#52506)

2026-09-01 10:26:43 -04:00

__init__.py

[V1] AsyncLLM Implementation (#9826)

2024-11-11 23:05:38 +00:00

test_kv_cache_spec_registry.py

[5/N][KV-Cache Layout Refactor] Backend-published KV packing via customize_spec (#51704)

2026-08-14 09:47:55 -04:00

test_oracle.py

[Docs] Clean up speculators docs (#34065)

2026-02-18 13:48:11 -08:00

test_outputs.py

[Feature][DSpark]: Logprobs adaptive verification (#52242)

2026-08-25 10:22:33 -04:00

test_request.py

feat(frontend): session id plumbing into requests (#48048)

2026-08-03 22:18:58 +00:00

test_serial_utils.py

[Bugfix] Don't reuse engine core payload buffer while zmq is sending it (#50053)

2026-07-28 16:09:28 +00:00

test_tensor_ipc_queue.py

[CI] Stabilize tensor IPC multiprocessing tests (#50804)

2026-08-12 10:16:43 +08:00

utils.py

Convert formatting to use ruff instead of yapf + isort (#26247)

2025-10-05 07:06:22 -07:00