weight-only-ffn
Small-M BF16-activation FFN regions with static W4 or W8 weights for Blackwell SM120/SM121.
Available functions:
quantize_w4_weight_bf16dequantize_w4_weight_bf16quantize_w8_weight_bf16dequantize_w8_weight_bf16w4a16_linear_bf16w8a16_linear_bf16w4a16_swiglu_ffn_bf16w4a16_geglu_ffn_bf16w4a16_gelu_ffn_bf16w8a16_swiglu_ffn_bf16w8a16_geglu_ffn_bf16w8a16_gelu_ffn_bf16
from kernels import get_kernel
ops = get_kernel(
"flashrt/weight-only-ffn",
version=1,
trust_remote_code=True,
)
packed, scales = ops.quantize_w8_weight_bf16(weight_bf16)
y = ops.w8a16_linear_bf16(x_bf16, packed, scales)
Production auto dispatch supports qualified M=1..4 shapes and rejects known
slow regions based on row count and weight dimensions.
Weights are prepared once; activations remain BF16 throughout the public
contract.
- Downloads last month
- 6
cuda
native-cuda
flashrt
blackwell
weight-only
int8
nvfp4
transformers
apache-2.0
Supported hardwares new
CUDA
- OS
- linux
- Arch
- x86_64
- Kernel Builder
- 19aaa64


