[Bugfix][CI/Build] Fix Plamo2 HF runner crash on transformers v5 (_tied_weights_keys list→dict) (#44239)

Nikhil Kulkarni committed 3 months ago

[Bugfix][Spec Decode] Select earliest-completing stop string in check_stop_strings (#49391)

Junpu Yu committed 3 months ago

[Bugfix] handle grammar compilation failures to avoid engine crash (#47312)

zhrrr committed 3 months ago

[Bugfix][Core] shm_broadcast: bound idle reader waits and release read slots (#45224)

chaeminlim-mb committed 3 months ago

[MRV2][Spec Decode] Avoid rejection sampler OOM by chunking (#48630)

Michael Goin committed 3 months ago

[Core] Simplify KVBlockZeroer index tensor handling (#48399)

Nick Hill committed 3 months ago

[MooncakeStore] Re-derive full external hits on stored boundaries (#49481)

Dao007forever committed 3 months ago

[Perf][KVConnector][Mooncake] Vectorize prepare_value on the KV load path (#48531)

Summer Yang committed 3 months ago

[CPU][Docs] Update docs and dockerfile for s390x (#49523)

Rehan Khan committed 3 months ago

[Misc] Use VLLMValidationError in chat_utils content-part validation (#49217)

Umut Polat committed 3 months ago

[Bugfix] Fix DeepSeek-V4 DSpark draft shared-expert padding for TP > 8 (#49415)

Mike G committed 3 months ago

[Performance][Model] Avoid transient Inkling result allocations (performance, and OOM prevention on smaller memory configurations) (#49487)

Mike G committed 3 months ago

[Bugfix][Model] Remove SciPy dependency from Inkling scale planning (#49485)

Mike G committed 3 months ago

[Docs] Re-add Reo.dev analytics beacon (#49474)

jcotant-inferact committed 3 months ago

[Bugfix] Make shared NVFP4 MoE scales writable (#49489)

Matej Sirovatka committed 3 months ago

Fused Shared Expert Support for AMD Quark DeepSeek-V4 Model Checkpoints (#48044)

Colin Z committed 3 months ago

[BugFix] Handle per-group prefix-hit divergence for hybrid models with KV connector (#48425)

Nick Hill committed 3 months ago

Add quantization label automation (#49492)

Michael Goin committed 3 months ago

[Core][DSV4] Compact MXFP4 indexer KV cache and packed group overlays (#48993)

Summer Yang committed 3 months ago

[Bugfix] Exclude location-derived path vars from torch.compile cache factors (#47573)

Nils Matteson committed 3 months ago