Zhiyu 431535b522

Enable modelopt gemma3 nvfp4/fp8, make workflow more robust (#22771) ...

Signed-off-by: Zhiyu Cheng <zhiyuc@nvidia.com>
Signed-off-by: Michael Goin <mgoin64@gmail.com>
Co-authored-by: Michael Goin <mgoin64@gmail.com>

2025-09-19 22:40:33 +00:00

..

attention

Optimize triton unified attention performance for sliding window attention (#24390)

2025-09-19 13:07:26 -06:00

core

[Kernel][Performance] Add Triton kernel for Qwen3-VL interleaved MRoPE (#25055)

2025-09-19 10:27:49 +00:00

mamba

fix some typos (#24616)

2025-09-11 10:48:46 -07:00

moe

Enable modelopt gemma3 nvfp4/fp8, make workflow more robust (#22771)

2025-09-19 22:40:33 +00:00

quantization

[Kernel] Delegate construction of FusedMoEQuantConfig to FusedMoEMethodBase subclasses (#22537)

2025-09-17 17:43:31 -06:00

__init__.py

[CI/Build] Move test_utils.py to tests/utils.py (#4425)

2024-05-13 23:50:09 +09:00

allclose_default.py

[Misc] Add SPDX-FileCopyrightText (#19100)

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

quant_utils.py

[Refactor] Remove Duplicate per_block_cast_to_fp8, Remove Dependencies of DeepGEMM (#21787)

2025-08-01 01:13:27 +00:00

test_apply_repetition_penalties.py

[BUG] Fix #20484. Support empty sequence in cuda penalty kernel (#20491)

2025-07-05 19:38:02 -07:00

test_flex_attention.py

Updates to Flex + VLLm integration (#21416)

2025-08-25 09:32:42 -04:00

test_fused_quant_activation.py

[Misc] Add SPDX-FileCopyrightText (#19100)

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

test_onednn.py

[CPU] Refactor CPU unquantized linear (#24150)

2025-09-04 14:28:45 +08:00

test_shuffle_rows.py

[Bugfix] Fix CUDA arch flags for MoE permute (#21426)

2025-07-24 03:23:59 -07:00

test_triton_flash_attention.py

[Misc] Add SPDX-FileCopyrightText (#19100)

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

utils.py

[Doc]: fixing typos to improve docs (#24480)

2025-09-08 23:06:04 -07:00