Thomas Parnell 6cbc4d4bea [Model] Add ModelConfig class for GraniteMoeHybrid to override default max_seq_len_to_capture (#20923) ...

Signed-off-by: Thomas Parnell <tpa@zurich.ibm.com>

2025-07-15 19:19:10 -07:00

..

guided_decoding

[V0][V1][Core] Add outlines integration for V1, and update V0 integration. (#15975)

2025-07-10 15:30:26 -04:00

layers

[BugFix] fix 3 issues: (1) using metadata for causal-conv1d, (2) indexing overflow in v1 vLLM, and (3) init_states in v0 (#20838)

2025-07-15 16:08:26 -04:00

model_loader

Enabled BnB NF4 inference on Gaudi (#20172)

2025-07-14 20:26:08 -07:00

models

[Model] Add ModelConfig class for GraniteMoeHybrid to override default max_seq_len_to_capture (#20923)

2025-07-15 19:19:10 -07:00

__init__.py

[Misc] Add SPDX-FileCopyrightText (#19100)

2025-06-03 11:20:17 -07:00

custom_op.py

[Fix][torch.compile] Enable custom ops by default when Inductor off (#20102)

2025-06-27 09:00:42 -06:00

parameter.py

[Misc] Add SPDX-FileCopyrightText (#19100)

2025-06-03 11:20:17 -07:00

pooling_metadata.py

[Misc] Add SPDX-FileCopyrightText (#19100)

2025-06-03 11:20:17 -07:00

sampling_metadata.py

[Misc] Add SPDX-FileCopyrightText (#19100)

2025-06-03 11:20:17 -07:00

utils.py

[Quant] [Bugfix] Fix quantization config matching with hf_to_vllm_mapper (#20046)

2025-07-01 19:20:34 +09:00