Michael Goin 09e4576f65 [Kernel] Add non-gated support for NVFP4 CUTLASS MoE (#37320) ...

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

2026-03-17 18:12:04 -04:00

..

moe-refactor

[Kernel] Add non-gated support for NVFP4 CUTLASS MoE (#37320)

2026-03-17 18:12:04 -04:00

moe-refactor-dp-ep

[UX] Add --moe-backend arg for explicit kernel selection (#33807)

2026-02-25 17:44:44 -08:00

DeepSeek-R1-DP.yaml

[CI] Bump num_speculative_tokens to 3 in nightly DeepSeek tests (#35882)

2026-03-03 09:26:44 -08:00

DeepSeek-R1-TP.yaml

[CI] Bump num_speculative_tokens to 3 in nightly DeepSeek tests (#35882)

2026-03-03 09:26:44 -08:00

DeepSeek-V2-Lite-Instruct-FP8.yaml

[CI] Generalize gsm8k test args and add Qwen3-Next MTP B200 test (#30723)

2025-12-16 14:28:34 -05:00

DeepSeek-V3.2-DP.yaml

[CI] Bump num_speculative_tokens to 3 in nightly DeepSeek tests (#35882)

2026-03-03 09:26:44 -08:00

DeepSeek-V3.2-TP.yaml

[CI] Bump num_speculative_tokens to 3 in nightly DeepSeek tests (#35882)

2026-03-03 09:26:44 -08:00

Llama-3-8B-Instruct-nonuniform-CT.yaml

[CI] Generalize gsm8k test args and add Qwen3-Next MTP B200 test (#30723)

2025-12-16 14:28:34 -05:00

Llama-3.2-1B-Instruct-INT8-CT.yaml

[CI] Generalize gsm8k test args and add Qwen3-Next MTP B200 test (#30723)

2025-12-16 14:28:34 -05:00

models-blackwell.txt

[CI] Add Qwen3-Next-FP8 to Blackwell model tests (#31049)

2025-12-23 17:21:50 -08:00

models-h200.txt

[CI] Add DeepSeek V3.2 nightly eval (#33566)

2026-02-02 16:10:02 +00:00

models-mi355.txt

Enabling some B200-specific tests on MI355 (#35253)

2026-03-06 19:27:20 +00:00

models-small.txt

[Quant] Support MXFP4 W4A16 for compressed-tensors MoE models (#32285)

2026-01-15 07:25:55 -08:00

Qwen1.5-MoE-W4A16-CT.yaml

[CI] Generalize gsm8k test args and add Qwen3-Next MTP B200 test (#30723)

2025-12-16 14:28:34 -05:00

Qwen2.5-VL-3B-Instruct-FP8-dynamic.yaml

[CI] Generalize gsm8k test args and add Qwen3-Next MTP B200 test (#30723)

2025-12-16 14:28:34 -05:00

Qwen3-0.6B-FP8.yaml

[CI] Generalize gsm8k test args and add Qwen3-Next MTP B200 test (#30723)

2025-12-16 14:28:34 -05:00

Qwen3-30B-A3B-MXFP4A16.yaml

[Quant] Support MXFP4 W4A16 for compressed-tensors MoE models (#32285)

2026-01-15 07:25:55 -08:00

Qwen3-30B-A3B-NVFP4.yaml

[CI] Generalize gsm8k test args and add Qwen3-Next MTP B200 test (#30723)

2025-12-16 14:28:34 -05:00

Qwen3-Next-80B-A3B-NVFP4-EP2.yaml

[UX] Add --moe-backend arg for explicit kernel selection (#33807)

2026-02-25 17:44:44 -08:00

Qwen3-Next-FP8-EP2_MI355.yaml

Enabling some B200-specific tests on MI355 (#35253)

2026-03-06 19:27:20 +00:00

Qwen3-Next-FP8-EP2.yaml

[UX] Add --moe-backend arg for explicit kernel selection (#33807)

2026-02-25 17:44:44 -08:00