Fused linear GELU GRN linear
Fused MLP pathway for ConvNextV2: expansion linear (dim -> 4*dim), GELU activation, Global Response Normalization (GRN) with L2 norm and channel-wise normalization, and projection linear (4*dim -> dim). Input/output in NHWC format.
Reported evidence · last observed 2026-07-10. Reported by source; not independently reproduced. No public source. License unknown.
Current records
Workloadsuite of 14 cases · mean latencydim = 128 · hidden_dim = 51214 cases
Not measured on H100 for this workload. Challenges →
145.5µs74% SOL1.00×Reported · license unknown · no source2026-07-10
Measured exactly what you asked. Reported by source; not independently reproduced.
246.5µs74% SOL1.02×Reported · license unknown · no source2026-07-16
Measured exactly what you asked. Reported by source; not independently reproduced.
346.6µs74% SOL1.02×Reported · license unknown · no source2026-07-10
Measured exactly what you asked. Reported by source; not independently reproduced.
447.6µs73% SOL1.05×Reported · license unknown · no source2026-08-15
Measured exactly what you asked. Reported by source; not independently reproduced.
547.9µs73% SOL1.05×Reported · license unknown · no source2026-08-14
Measured exactly what you asked. Reported by source; not independently reproduced.
649.0µs73% SOL1.08×Reported · license unknown · no source2026-07-10
Measured exactly what you asked. Reported by source; not independently reproduced.
749.3µs72% SOL1.08×Reported · license unknown · no source2026-07-10
Measured exactly what you asked. Reported by source; not independently reproduced.
854.8µs70% SOL1.20×Reported · license unknown · no source2026-07-27
Measured exactly what you asked. Reported by source; not independently reproduced.
955.0µs70% SOL1.21×Reported · license unknown · no source2026-08-14
Measured exactly what you asked. Reported by source; not independently reproduced.
1055.3µs70% SOL1.21×Reported · license unknown · no source2026-07-20
Measured exactly what you asked. Reported by source; not independently reproduced.
1157.9µs69% SOL1.27×Reported · license unknown · no source2026-07-10
Measured exactly what you asked. Reported by source; not independently reproduced.
1261.1µs67% SOL1.34×Reported · license unknown · no source2026-08-23
Measured exactly what you asked. Reported by source; not independently reproduced.
1361.7µs67% SOL1.35×Reported · license unknown · no source2026-08-12
Measured exactly what you asked. Reported by source; not independently reproduced.
1461.8µs67% SOL1.36×Reported · license unknown · no source2026-08-11
Measured exactly what you asked. Reported by source; not independently reproduced.
1564.0µs66% SOL1.41×Reported · license unknown · no source2026-07-10
Measured exactly what you asked. Reported by source; not independently reproduced.
1670.2µs64% SOL1.54×Reported · license unknown · no source2026-08-10
Measured exactly what you asked. Reported by source; not independently reproduced.
1786.1µs59% SOL1.89×Reported · license unknown · no source2026-08-09
Measured exactly what you asked. Reported by source; not independently reproduced.
1886.5µs58% SOL1.90×Reported · license unknown · no source2026-08-10
Measured exactly what you asked. Reported by source; not independently reproduced.
1989.8µs57% SOL1.97×Reported · license unknown · no source2026-08-07
Measured exactly what you asked. Reported by source; not independently reproduced.
20108.1µs53% SOL2.38×Reported · license unknown · no source2026-07-24
Measured exactly what you asked. Reported by source; not independently reproduced.
21219.1µs35% SOL4.81×Reported · license unknown · no source2026-07-10
Measured exactly what you asked. Reported by source; not independently reproduced.