Compare runs
cohort sha256:d54c24b8a75223e0… · source-native profile
ranking-v1
no ranks: not comparable
Comparison identity
operation
grouped-gemm-fp8-fp4-m-masked-g4-n2048-k2048
workload
max_m = 512 · fp32/fp8_e4m3 · 19c6e666
protocol
flashinfer-bench · 7d44d206
environment
NVIDIA B200 · 83c9b6e7
correctness policy
64fb869e
metric
latency mean (ns)
Context · 10 fields›
architecture
sm_100
CUDA
13.0
driver
unknown
framework
pytorch 2.11.0+cu130
samples
unknown
evidence
reported
license
Apache-2.0
source
available
status
passed
observed
2026-06-06
No winner: these runs didn't measure the same thing. The rows above show what differs. Why comparable?
Add run ›
Export ›
Markdown CSV JSON
deepgemm_m_grouped_fp8_fp4_masked_g4_n2048_k2048 · sha256:1af005b6360aa… · observed 2026-06-06