MoE FP8 block scale ds routing topk8 ng8 kg4 e32 h7168 i2048
FP8 block scale MoE operation. Routing and two grouped-GEMM included.
Reported evidence · last observed 2026-07-17. Reported by source; not independently reproduced. No public source. License unknown.
Current records
Workloadsuite of 19 cases · mean latency38 cases
Not measured on H100 for this workload. Challenges →
1271.3µs97% SOL1.00×Reported · license unknown · no source2026-07-17
Measured exactly what you asked. Reported by source; not independently reproduced.
2273.8µs97% SOL1.01×Reported · license unknown · no source2026-07-29
Measured exactly what you asked. Reported by source; not independently reproduced.
3281.1µs97% SOL1.04×Reported · license unknown · no source2026-07-13
Measured exactly what you asked. Reported by source; not independently reproduced.
4301.2µs96% SOL1.11×Reported · license unknown · no source2026-08-21
Measured exactly what you asked. Reported by source; not independently reproduced.
5307.4µs96% SOL1.13×Reported · license unknown · no source2026-07-13
Measured exactly what you asked. Reported by source; not independently reproduced.
6313.5µs96% SOL1.16×Reported · license unknown · no source2026-08-20
Measured exactly what you asked. Reported by source; not independently reproduced.
7314.4µs96% SOL1.16×Reported · license unknown · no source2026-08-18
Measured exactly what you asked. Reported by source; not independently reproduced.
8316.0µs96% SOL1.16×Reported · license unknown · no source2026-08-14
Measured exactly what you asked. Reported by source; not independently reproduced.
9317.0µs96% SOL1.17×Reported · license unknown · no source2026-08-21
Measured exactly what you asked. Reported by source; not independently reproduced.
10327.9µs96% SOL1.21×Reported · license unknown · no source2026-08-15
Measured exactly what you asked. Reported by source; not independently reproduced.
11331.3µs96% SOL1.22×Reported · license unknown · no source2026-07-13
Measured exactly what you asked. Reported by source; not independently reproduced.
12333.6µs95% SOL1.23×Reported · license unknown · no source2026-08-11
Measured exactly what you asked. Reported by source; not independently reproduced.
13340.8µs95% SOL1.26×Reported · license unknown · no source2026-07-13
Measured exactly what you asked. Reported by source; not independently reproduced.
14347.9µs95% SOL1.28×Reported · license unknown · no source2026-08-09
Measured exactly what you asked. Reported by source; not independently reproduced.
15365.3µs95% SOL1.35×Reported · license unknown · no source2026-08-06
Measured exactly what you asked. Reported by source; not independently reproduced.
16376.6µs95% SOL1.39×Reported · license unknown · no source2026-08-05
Measured exactly what you asked. Reported by source; not independently reproduced.
17418.8µs94% SOL1.54×Reported · license unknown · no source2026-08-06
Measured exactly what you asked. Reported by source; not independently reproduced.
18456.0µs93% SOL1.68×Reported · license unknown · no source2026-08-02
Measured exactly what you asked. Reported by source; not independently reproduced.
19459.9µs93% SOL1.70×Reported · license unknown · no source2026-08-01
Measured exactly what you asked. Reported by source; not independently reproduced.
20460.0µs93% SOL1.70×Reported · license unknown · no source2026-08-01
Measured exactly what you asked. Reported by source; not independently reproduced.
21466.4µs93% SOL1.72×Reported · license unknown · no source2026-08-02
Measured exactly what you asked. Reported by source; not independently reproduced.
22481.5µs93% SOL1.78×Reported · license unknown · no source2026-08-04
Measured exactly what you asked. Reported by source; not independently reproduced.
23485.3µs93% SOL1.79×Reported · license unknown · no source2026-07-13
Measured exactly what you asked. Reported by source; not independently reproduced.
24492.2µs92% SOL1.81×Reported · license unknown · no source2026-08-01
Measured exactly what you asked. Reported by source; not independently reproduced.
25521.3µs92% SOL1.92×Reported · license unknown · no source2026-07-30
Measured exactly what you asked. Reported by source; not independently reproduced.
26523.1µs92% SOL1.93×Reported · license unknown · no source2026-07-30
Measured exactly what you asked. Reported by source; not independently reproduced.
27524.7µs92% SOL1.93×Reported · license unknown · no source2026-08-01
Measured exactly what you asked. Reported by source; not independently reproduced.
28525.9µs92% SOL1.94×Reported · license unknown · no source2026-07-31
Measured exactly what you asked. Reported by source; not independently reproduced.
29535.0µs92% SOL1.97×Reported · license unknown · no source2026-07-30
Measured exactly what you asked. Reported by source; not independently reproduced.
30546.9µs92% SOL2.02×Reported · license unknown · no source2026-07-24
Measured exactly what you asked. Reported by source; not independently reproduced.
31547.1µs92% SOL2.02×Reported · license unknown · no source2026-07-24
Measured exactly what you asked. Reported by source; not independently reproduced.
32593.8µs90% SOL2.19×Reported · license unknown · no source2026-08-03
Measured exactly what you asked. Reported by source; not independently reproduced.
33648.9µs89% SOL2.39×Reported · license unknown · no source2026-07-26
Measured exactly what you asked. Reported by source; not independently reproduced.
34715.2µs87% SOL2.64×Reported · license unknown · no source2026-07-18
Measured exactly what you asked. Reported by source; not independently reproduced.
35861.5µs86% SOL3.18×Reported · license unknown · no source2026-07-24
Measured exactly what you asked. Reported by source; not independently reproduced.
36962.3µs85% SOL3.55×Reported · license unknown · no source2026-07-24
Measured exactly what you asked. Reported by source; not independently reproduced.
371.07ms83% SOL3.95×Reported · license unknown · no source2026-07-24
Measured exactly what you asked. Reported by source; not independently reproduced.
381.09ms83% SOL4.00×Reported · license unknown · no source2026-07-28
Measured exactly what you asked. Reported by source; not independently reproduced.
391.43ms78% SOL5.26×Reported · license unknown · no source2026-08-02
Measured exactly what you asked. Reported by source; not independently reproduced.
401.80ms73% SOL6.64×Reported · license unknown · no source2026-08-04
Measured exactly what you asked. Reported by source; not independently reproduced.
412.29ms68% SOL8.45×Reported · license unknown · no source2026-07-13
Measured exactly what you asked. Reported by source; not independently reproduced.
4211.8ms30% SOL43.48×Reported · license unknown · no source2026-07-14
Measured exactly what you asked. Reported by source; not independently reproduced.
4331.3ms16% SOL115.54×Reported · license unknown · no source2026-07-14
Measured exactly what you asked. Reported by source; not independently reproduced.
Implementations
Show all 79 implementations ›Showing all 79 implementations ⌄
Semantics
Also indexed as 020-moe-fp8-block-scale-ds-routing-topk8-ng8-kg4-e32-h7168-i2048 · reviewed as the same computation; every definition's cohorts are shown here, ranked separately.