Skip to content

Search Models Records GPUs Docs

More

Records GPUs Docs

Kernels Serving

⏎

Feed Serving Projects Challenges Contribute Sources Methodology Legal GitHub Atom llms.txta46b010

Kernels Serving

⏎

gpu B200✕

16 operations match

Pick one to compare its implementations; your filters carry over

Operation

Family

Published runs

Last run

Attention softmax dropout value matmul backward128 on B200 · best 238.7 µsattention1282026-08-24 Attention softmax with softcapping and dropout26 on B200 · best 7.51 µsattention262026-08-24 Fused attention QK matmul scale mask softmax backward39 on B200 · best 137.0 µsattention392026-08-24 Masked softmax with attention dropout backward22 on B200 · best 24.3 µsattention222026-08-19 Conv3d Min Softmaxnone on B200conv42026-03-05 Conv3d Softmax MaxPool MaxPoolnone on B200conv42026-03-05 ConvTranspose2d Softmax BiasAdd Scaling Sigmoidnone on B200conv42026-03-05 ConvTranspose3d AvgPool Clamp Softmax Multiplynone on B200conv42026-03-05 ConvTranspose3d MaxPool Softmax Subtract Swish Maxnone on B200conv42026-03-05 ConvTranspose3d Softmax Sigmoidnone on B200conv42026-03-05 Gemm BatchNorm Scaling Softmaxnone on B200gemm42026-03-05 Matmul Dropout Softmaxnone on B200gemm42026-03-05 Matmul GELU Softmaxnone on B200gemm42026-03-05 LogSoftmaxnone on B200softmax42026-03-05 Softmaxnone on B200softmax482025-04-30 Softmaxnone on B200softmax42026-03-05

Ranked only against runs that measured the same thing.