Cross attention text video conditioning backward
Backward pass for cross-attention text-video conditioning. Computes gradients through output projection, attention mechanism (softmax, Q@K^T), and Q/K/V linear projections. Queries from video latents, keys/values from text embeddings.
Reported evidence · last observed 2026-08-18. Reported by source; not independently reproduced. No public source. License unknown.
Current records
Workloadsuite of 16 cases · mean latencyhead_dim = 64 · num_heads = 16 · hidden_size = 102416 cases
Not measured on H100 for this workload. Challenges →
1114.5µs75% SOL1.00×Reported · license unknown · no source2026-08-18
Measured exactly what you asked. Reported by source; not independently reproduced.
2144.7µs70% SOL1.26×Reported · license unknown · no source2026-07-26
Measured exactly what you asked. Reported by source; not independently reproduced.
3150.3µs69% SOL1.31×Reported · license unknown · no source2026-07-17
Measured exactly what you asked. Reported by source; not independently reproduced.
4154.4µs68% SOL1.35×Reported · license unknown · no source2026-08-04
Measured exactly what you asked. Reported by source; not independently reproduced.
5168.7µs66% SOL1.47×Reported · license unknown · no source2026-07-10
Measured exactly what you asked. Reported by source; not independently reproduced.
6169.4µs66% SOL1.48×Reported · license unknown · no source2026-07-10
Measured exactly what you asked. Reported by source; not independently reproduced.
7178.5µs65% SOL1.56×Reported · license unknown · no source2026-07-10
Measured exactly what you asked. Reported by source; not independently reproduced.
8182.0µs64% SOL1.59×Reported · license unknown · no source2026-08-18
Measured exactly what you asked. Reported by source; not independently reproduced.
9182.2µs64% SOL1.59×Reported · license unknown · no source2026-08-18
Measured exactly what you asked. Reported by source; not independently reproduced.
10182.4µs64% SOL1.59×Reported · license unknown · no source2026-08-16
Measured exactly what you asked. Reported by source; not independently reproduced.
11183.4µs64% SOL1.60×Reported · license unknown · no source2026-08-24
Measured exactly what you asked. Reported by source; not independently reproduced.
12209.8µs61% SOL1.83×Reported · license unknown · no source2026-08-23
Measured exactly what you asked. Reported by source; not independently reproduced.
13214.4µs60% SOL1.87×Reported · license unknown · no source2026-08-22
Measured exactly what you asked. Reported by source; not independently reproduced.
14216.4µs60% SOL1.89×Reported · license unknown · no source2026-07-21
Measured exactly what you asked. Reported by source; not independently reproduced.
15235.0µs58% SOL2.05×Reported · license unknown · no source2026-07-14
Measured exactly what you asked. Reported by source; not independently reproduced.
16242.9µs57% SOL2.12×Reported · license unknown · no source2026-08-14
Measured exactly what you asked. Reported by source; not independently reproduced.
17263.9µs55% SOL2.30×Reported · license unknown · no source2026-08-10
Measured exactly what you asked. Reported by source; not independently reproduced.
18296.2µs52% SOL2.59×Reported · license unknown · no source2026-07-10
Measured exactly what you asked. Reported by source; not independently reproduced.
19331.5µs49% SOL2.89×Reported · license unknown · no source2026-07-10
Measured exactly what you asked. Reported by source; not independently reproduced.
20384.9µs46% SOL3.36×Reported · license unknown · no source2026-08-22
Measured exactly what you asked. Reported by source; not independently reproduced.
21408.2µs44% SOL3.56×Reported · license unknown · no source2026-08-06
Measured exactly what you asked. Reported by source; not independently reproduced.
222.59ms11% SOL22.65×Reported · license unknown · no source2026-08-09
Measured exactly what you asked. Reported by source; not independently reproduced.