Michael Goin 5e5c8e091e [Quant][Perf] Use moe_wna16 kernel by default for MoEs with many experts (#13236) ...

Signed-off-by: mgoin <mgoin64@gmail.com>

2025-02-14 12:53:42 -08:00

..

fused_moe

Optimize moe_align_block_size for deepseek_v3 (#12850)

2025-02-13 18:43:37 -05:00

mamba

Add Bamba Model (#10909)

2025-02-06 15:22:42 -08:00

quantization

[Quant][Perf] Use moe_wna16 kernel by default for MoEs with many experts (#13236)

2025-02-14 12:53:42 -08:00

__init__.py

Change the name to vLLM (#150)

2023-06-17 03:07:40 -07:00

activation.py

[Misc] Add SPDX-License-Identifier headers to python source files (#12628)

2025-02-02 11:58:18 -08:00

layernorm.py

[Misc] Add SPDX-License-Identifier headers to python source files (#12628)

2025-02-02 11:58:18 -08:00

linear.py

[Bugfix] Allow fallback to AWQ from AWQMarlin at per-layer granularity (#13119)

2025-02-12 09:19:53 -08:00

logits_processor.py

[CORE] [QUANT] Support for GPTQModel's dynamic quantization per module override/control (#7086)

2025-02-12 09:19:43 -08:00

pooler.py

[Misc] Add SPDX-License-Identifier headers to python source files (#12628)

2025-02-02 11:58:18 -08:00

rejection_sampler.py

[Misc] Add SPDX-License-Identifier headers to python source files (#12628)

2025-02-02 11:58:18 -08:00

resampler.py

[Misc] Add SPDX-License-Identifier headers to python source files (#12628)

2025-02-02 11:58:18 -08:00

rotary_embedding.py

[Hardware][Intel-Gaudi] Enable long-contexts + LoRA support for Intel Gaudi (#12812)

2025-02-08 17:15:30 +08:00

sampler.py

[Misc] Add SPDX-License-Identifier headers to python source files (#12628)

2025-02-02 11:58:18 -08:00

spec_decode_base_sampler.py

[Misc] Add SPDX-License-Identifier headers to python source files (#12628)

2025-02-02 11:58:18 -08:00

typical_acceptance_sampler.py

[Misc] Add SPDX-License-Identifier headers to python source files (#12628)

2025-02-02 11:58:18 -08:00

utils.py

[Misc] Add SPDX-License-Identifier headers to python source files (#12628)

2025-02-02 11:58:18 -08:00

vocab_parallel_embedding.py

[CORE] [QUANT] Support for GPTQModel's dynamic quantization per module override/control (#7086)

2025-02-12 09:19:43 -08:00