GQA ragged prefill causal h32 kv4 d128
Batched Grouped Query Attention prefill with ragged (variable-length) inputs. Causal mask is applied. Captured from Qwen3-30B-A3B during total prefill.
Reported evidence · last observed 2026-08-21. Reported by source; not independently reproduced. No public source. License unknown.
Current records
Workloadsuite of 15 cases · mean latency17 cases
Not measured on H100 for this workload. Challenges →
17.00µs94% SOL1.00×Reported · license unknown · no source2026-08-21
Measured exactly what you asked. Reported by source; not independently reproduced.
27.31µs93% SOL1.04×Reported · license unknown · no source2026-07-30
Measured exactly what you asked. Reported by source; not independently reproduced.
37.50µs93% SOL1.07×Reported · license unknown · no source2026-08-06
Measured exactly what you asked. Reported by source; not independently reproduced.
47.71µs93% SOL1.10×Reported · license unknown · no source2026-07-10
Measured exactly what you asked. Reported by source; not independently reproduced.
57.74µs93% SOL1.11×Reported · license unknown · no source2026-08-03
Measured exactly what you asked. Reported by source; not independently reproduced.
67.86µs93% SOL1.12×Reported · license unknown · no source2026-08-21
Measured exactly what you asked. Reported by source; not independently reproduced.
78.13µs92% SOL1.16×Reported · license unknown · no source2026-08-21
Measured exactly what you asked. Reported by source; not independently reproduced.
88.14µs92% SOL1.16×Reported · license unknown · no source2026-07-31
Measured exactly what you asked. Reported by source; not independently reproduced.
98.20µs92% SOL1.17×Reported · license unknown · no source2026-07-18
Measured exactly what you asked. Reported by source; not independently reproduced.
108.25µs92% SOL1.18×Reported · license unknown · no source2026-08-01
Measured exactly what you asked. Reported by source; not independently reproduced.
118.34µs92% SOL1.19×Reported · license unknown · no source2026-07-28
Measured exactly what you asked. Reported by source; not independently reproduced.
128.34µs92% SOL1.19×Reported · license unknown · no source2026-07-28
Measured exactly what you asked. Reported by source; not independently reproduced.
138.37µs92% SOL1.20×Reported · license unknown · no source2026-08-03
Measured exactly what you asked. Reported by source; not independently reproduced.
148.48µs92% SOL1.21×Reported · license unknown · no source2026-08-02
Measured exactly what you asked. Reported by source; not independently reproduced.
158.61µs92% SOL1.23×Reported · license unknown · no source2026-08-05
Measured exactly what you asked. Reported by source; not independently reproduced.
168.94µs92% SOL1.28×Reported · license unknown · no source2026-07-30
Measured exactly what you asked. Reported by source; not independently reproduced.
179.08µs91% SOL1.30×Reported · license unknown · no source2026-07-10
Measured exactly what you asked. Reported by source; not independently reproduced.
189.31µs91% SOL1.33×Reported · license unknown · no source2026-07-16
Measured exactly what you asked. Reported by source; not independently reproduced.
199.35µs91% SOL1.34×Reported · license unknown · no source2026-08-03
Measured exactly what you asked. Reported by source; not independently reproduced.
209.79µs91% SOL1.40×Reported · license unknown · no source2026-07-27
Measured exactly what you asked. Reported by source; not independently reproduced.
219.91µs90% SOL1.42×Reported · license unknown · no source2026-07-29
Measured exactly what you asked. Reported by source; not independently reproduced.
2211.0µs90% SOL1.57×Reported · license unknown · no source2026-07-24
Measured exactly what you asked. Reported by source; not independently reproduced.
2311.6µs89% SOL1.65×Reported · license unknown · no source2026-07-10
Measured exactly what you asked. Reported by source; not independently reproduced.
2418.4µs83% SOL2.63×Reported · license unknown · no source2026-07-28
Measured exactly what you asked. Reported by source; not independently reproduced.
2518.8µs83% SOL2.69×Reported · license unknown · no source2026-07-19
Measured exactly what you asked. Reported by source; not independently reproduced.
2625.0µs78% SOL3.57×Reported · license unknown · no source2026-07-10
Measured exactly what you asked. Reported by source; not independently reproduced.
2725.5µs78% SOL3.65×Reported · license unknown · no source2026-07-10
Measured exactly what you asked. Reported by source; not independently reproduced.
Implementations
Show all 48 implementations ›Showing all 48 implementations ⌄
Semantics
Also indexed as 016-gqa-ragged-prefill-causal-h32-kv4-d128 · reviewed as the same computation; every definition's cohorts are shown here, ranked separately.