Figure 5a reports forward attention throughput on A100 80GB SXM4; PyTorch is marked OOM at sequence length 16k. The panel covers non-causal attention with head dimension 64.

Figure 5a reports forward attention throughput on A100 80GB SXM4; PyTorch is marked OOM at sequence length 16k. The panel covers non-causal attention with head dimension 64. · CiteArk

Measurement results

0/29 assessed

0 supported0 disputed or conflicting29 not assessed

The complete result group stays with its claim. Assessed rows come first; their judgments do not replace the whole-claim conclusion. Expand a row for conditions, sources, and evidence history.

Each row is a measurement with its reported value, latest available observation, and cumulative evidence state.
Metric and conditions	Reported	Latest observation	Evidence judgment
method: PyTorch · sequenceLength: 512	29 TFLOPs/s	—	Not assessed
method: FlashAttention · sequenceLength: 512	91 TFLOPs/s	—	Not assessed
method: xFormers · sequenceLength: 512	89 TFLOPs/s	—	Not assessed
method: FlashAttention Triton · sequenceLength: 512	128 TFLOPs/s	—	Not assessed
method: FlashAttention-2 · sequenceLength: 512	178 TFLOPs/s	—	Not assessed
method: PyTorch · sequenceLength: 1k	34 TFLOPs/s	—	Not assessed
method: FlashAttention · sequenceLength: 1k	96 TFLOPs/s	—	Not assessed
method: xFormers · sequenceLength: 1k	94 TFLOPs/s	—	Not assessed
method: FlashAttention Triton · sequenceLength: 1k	141 TFLOPs/s	—	Not assessed
method: FlashAttention-2 · sequenceLength: 1k	191 TFLOPs/s	—	Not assessed
method: PyTorch · sequenceLength: 2k	35 TFLOPs/s	—	Not assessed
method: FlashAttention · sequenceLength: 2k	99 TFLOPs/s	—	Not assessed

Figure 5a reports forward attention throughput on A100 80GB SXM4; PyTorch is marked OOM at sequence length 16k. The panel covers non-causal attention with head dimension 64.

Reported and observed measurements

Assessments (0)

Runs (0)

Measurement results