Wing Lian b8358aa5ab [gemma4] use mixed Flash Attention and SDPA and add fused RMSNorm+RoPE Triton kernels (#3598)

2026-04-12 10:29:55 -04:00

..

test_gemma4_fused_rope.py

[gemma4] use mixed Flash Attention and SDPA and add fused RMSNorm+RoPE Triton kernels (#3598)

2026-04-12 10:29:55 -04:00

test_rms_norm_gated.py

liger support for qwen 3.5 and fused rmsnorm+gated (#3531) [skip ci]

2026-03-22 13:19:21 -04:00