Compare runs
cohort sha256:102304cfd31f5d2d… · source-native profile
ranking-v1
no ranks: not comparable
Comparison identity
operation
gqa-paged-prefill-causal-h5-kv1-d128-ps1
workload
total_q = 14355 · num_pages = 14522 · len_indptr = 44 · num_kv_indices = 14398 · bf16 · 578b5f1e
protocol
flashinfer-bench · 7d44d206
environment
NVIDIA GB200 · 1766500d
correctness policy
64fb869e
metric
latency mean (ns)
Context · 10 fields›
architecture
sm_100
CUDA
13.0
driver
unknown
framework
pytorch 2.9.1+cu130
samples
unknown
evidence
reproducible
license
Apache-2.0
source
available
status
passed
observed
2026-04-12
No winner: these runs didn't measure the same thing. The rows above show what differs. Why comparable?
Add run ›
Export ›
Markdown CSV JSON
flashinfer_wrapper_484d2c · sha256:37e34b9be5106… · observed 2026-04-12