Wei-Ming Chen 1c94e8dc7d Add NVFP4 KV 4-over-6 scale search (#45187) ...

Signed-off-by: weimingc <17592131+meenchen@users.noreply.github.com>

2026-08-07 13:22:27 -07:00

..

attention

Add NVFP4 KV 4-over-6 scale search (#45187)

2026-08-07 13:22:27 -07:00

core

fix: resolve silent request skipping in PRIORITY scheduling (#49206)

2026-08-06 22:21:21 +00:00

engine

[Feat] Support thinking_token_budget in Model Runner V2 (#46727)

2026-08-07 03:36:18 +00:00

executor

[Core] Explicitly manage torch CPU threads in workers (#49919)

2026-08-04 10:06:38 -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_offload

[Bugfix][KV Offload] Clean up resources after initialization failure (#51227)

2026-08-06 14:42:13 +03:00

metrics

[R3] Unify routed expert shape configuration (#50940)

2026-08-04 20:20:41 -07:00

pool

[Core] Move MRV1 late_interaction_runner.py out of MRV2 subtree (#48014)

2026-07-08 18:30:11 +00:00

sample

Enable gfx1250 ROCm architecture (#46516)

2026-07-31 01:41:12 +00:00

simple_kv_offload

[Feat][Core] Add disk offloading support to SimpleCPUOffloadConnector (#49644)

2026-08-07 15:37:47 +08:00

spec_decode

[Bugfix][Model Runner V2] Restore multimodal draft capability detection (#50417)

2026-08-03 14:18:14 -07:00

structured_output

[Bugfix] Restore structured output logger initialization (#49626)

2026-07-23 11:26:03 -07:00

worker

[MRv2 Feature] MR v2 weight offloading support (#51413)

2026-08-07 13:20:21 -06:00

__init__.py

…

cudagraph_dispatcher.py

[Model] Add DiffusionGemma Support (#45163)

2026-06-11 22:17:35 -07:00

kv_cache_interface.py

Add NVFP4 KV 4-over-6 scale search (#45187)

2026-08-07 13:22:27 -07:00

kv_cache_spec_registry.py

[KVCache] Support Pluggable KVCacheSpec (#37505)

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

outputs.py

[Bugfix][EC Connector] Don't stop an encoder-instance request before its images are encoded (#50275)

2026-08-05 05:53:58 +00:00

request.py

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

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

serial_utils.py

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

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

utils.py

[chore] log process manager shutdown with more details (#49437)

2026-07-31 11:55:09 -07:00