Nick Hillandziminghuang fcaa84efa7 [BugFix] Gate MRV2 mixed sparse-MLA warmup on max_num_seqs > 1 (#47050) ...

Signed-off-by: Nick Hill <nickhill123@gmail.com>
Co-authored-by: ziminghuang <ziminghuang@inferact.ai>

2026-06-30 16:31:27 +01:00

..

attention

[ROCm][Perf][MLA] Add AITER FlashAttention MLA prefill backend (ROCM_AITER_FA) (#45033)

2026-06-28 07:52:00 -07:00

core

[Model] Support Unlimited OCR (#46564)

2026-06-27 23:09:18 -07:00

cudagraph

[MM][Perf][CG] Support dual-path ViT full CUDA graph for DeepSeek-OCR (#43586)

2026-06-16 04:35:20 -07:00

determinism

[CI Test] Mark batch invariance test flaky (#46530)

2026-06-23 21:01:34 +00:00

distributed

[Core] Add prefill step cadence for better non-PD DP balancing (#44558)

2026-06-16 13:17:18 -07:00

e2e

[Model Runner V2] support mamba hybrid models align prefix cache (#42406)

2026-06-29 14:09:16 -07:00

ec_connector

[Docs] Fix broken local links (#42160)

2026-05-10 01:15:38 -07:00

engine

[Frontend] Split ServingRender into renderer and entrypoint. (#44285)

2026-06-23 11:19:09 +00:00

executor

[Core][AMD] Propagate shutdown timeout to MultiprocExecutor (#43154)

2026-06-12 15:13:31 -05:00

kv_connector

[Spec Decode] Fix hidden-state extraction block size for hybrid verifiers (#46301)

2026-06-30 08:19:51 -07:00

kv_offload

[KV Offload] Pass ScheduleEndContext to on_schedule_end hook (#46450)

2026-06-30 17:07:12 +03:00

logits_processors

[Bugfix][Reasoning] Fix thinking_token_budget not enforced on re-entry after forced end (#43757)

2026-06-30 01:04:25 +00:00

metrics

[V1][Metrics] Add MLA attention metrics for DeepSeek MFU estimation (#39457)

2026-06-12 14:28:40 +01:00

sample

[Platform] Replace torch.cuda.mem_get_info with torch.accelerator.get_memory_info (#44825)

2026-06-30 14:39:52 +08:00

shutdown

[ROCm][CI] Make tests/v1/shutdown an importable package (#47085)

2026-06-29 21:01:27 -05:00

simple_kv_offload

[SimpleCPUOffloadConnector] Fix remaining global→block conversions under PCP/DCP (#46394)

2026-06-24 23:05:24 -07:00

spec_decode

[Model Runner V2][Spec Decode] Implement block verification for rejection sampling (#46781)

2026-06-30 08:07:24 -07:00

streaming_input

[Core] Ensure memory is pinned prior to async h2d copy (#45424)

2026-06-20 20:02:24 -07:00

structured_output

[Bugfix] Reject structured outputs for diffusion decoders with a clear error (#45468)

2026-06-13 12:04:01 -07:00

tracing

[ROCm][CI] Use spawn around the threaded OTLP test (#47003)

2026-06-29 16:34:05 -05:00

worker

[BugFix] Gate MRV2 mixed sparse-MLA warmup on max_num_seqs > 1 (#47050)

2026-06-30 16:31:27 +01:00

__init__.py

[V1] AsyncLLM Implementation (#9826)

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

test_kv_cache_spec_registry.py

[KVCache] Support Pluggable KVCacheSpec (#37505)

2026-06-03 09:05:16 -07:00

test_oracle.py

[Docs] Clean up speculators docs (#34065)

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

test_outputs.py

[BugFix] Fix spec decoding max_tokens scheduling perf issue (#29542)

2025-11-28 20:52:23 +08:00

test_request.py

[Refactor] Rename WAITING_FOR_FSM to WAITING_FOR_STRUCTURED_OUTPUT_GRAMMAR (#38048)

2026-03-25 11:41:44 -04:00

test_serial_utils.py

Add tensor IPC transfer mechanism for multimodal data (#32104)

2026-03-21 20:10:20 +00:00

test_tensor_ipc_queue.py

[Test] Refactor hard coded device string in test files under compile/quantization/models/model_executor folders (#38901)

2026-04-15 11:02:35 +08:00

utils.py

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

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