Packs short training examples into fixed length sequences and reports the gradient weight each one actually receives
-
Updated
Aug 14, 2026 - TypeScript
#
sequence-packing
Here are 2 public repositories matching this topic...
Benchmark of sequence packing and length-aware batching for variable-length LLM requests, measuring compute waste from padding and the latency-efficiency trade-off between offline and online grouping policies.
python performance-engineering benchmark simulation latency scheduler inference batching systems throughput padding serving event-driven-simulation llm sharegpt heavy-tail sequence-packing
-
Updated
Jul 25, 2026 - Python
Improve this page
Add a description, image, and links to the sequence-packing topic page so that developers can more easily learn about it.
Add this topic to your repo
To associate your repository with the sequence-packing topic, visit your repo's landing page and select "manage topics."