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✕dtype bf16✕Clear filters

20 operations match

Pick one to compare its implementations; your filters carry over

Operation

Family

Published runs

Last run

GEMM MXFP4 n2048 k20489 on B200 · bf16 · best 273.6 µs · 9 with sourcegemm92026-06-06 GEMM MXFP4 n4096 k40969 on B200 · bf16 · best 271.3 µs · 9 with sourcegemm92026-06-06 GEMM MXFP4 n5120 k20489 on B200 · bf16 · best 274.7 µs · 9 with sourcegemm92026-06-06 GEMM MXFP4 out NVFP4 n2048 k20482 on B200 · bf16 · best 354.7 µs · 2 with sourcegemm22026-06-06 GEMM FP8 fp4 n2048 k2048none on B200 · bf16gemm32026-06-06 GEMM FP8 fp4 n4096 k4096none on B200 · bf16gemm32026-06-06 Gemm GroupNorm Swish Multiply Swishnone on B200 · bf16gemm42026-03-05 Gemm Max Subtract GELUnone on B200 · bf16gemm42026-03-05 Gemm Multiply LeakyReLUnone on B200 · bf16gemm42026-03-05 GEMM n256 k3072none on B200 · bf16gemmnone yet—GEMM n3072 k6144none on B200 · bf16gemmnone yet—GEMM n8192 k3072none on B200 · bf16gemmnone yet—GEMM NVFP4 n2048 k2048none on B200 · bf16gemm182026-06-06 GEMM NVFP4 n4096 k4096none on B200 · bf16gemm182026-06-06 Grouped GEMM FP8 fp4 m contiguous g4 n2048 k2048none on B200 · bf16gemm22026-06-06 Grouped GEMM FP8 fp4 m contiguous g4 n4096 k2048none on B200 · bf16gemm22026-06-06 Grouped GEMM FP8 fp4 m masked g4 n2048 k2048none on B200 · bf16gemm22026-06-06 Grouped GEMM FP8 fp4 m masked g4 n4096 k2048none on B200 · bf16gemm22026-06-06 Grouped GEMM mxfp8 MXFP4 g4 n2048 k2048none on B200 · bf16gemm22026-06-06 Grouped GEMM mxfp8 MXFP4 g4 n4096 k2048none on B200 · bf16gemm22026-06-06

Ranked only against runs that measured the same thing.