Source record

Published by Amp News (ampcode.com) (T1 source). The original is at https://ampcode.com/news/faster-deep-rush.

Pipeline notes

The summary and note below are generated by the signal pipeline — they are Beyond Desk’s reading, not quotations from the source.

Amp reports its deep and rush modes now deliver first tokens 87% faster and full responses 32% faster at p50, achieved mainly by switching to websockets for OpenAI communication plus prior internal rebuilds, with up to 40% end-to-end speedup on long-horizon tasks.

Vendor-published latency numbers with the specific transport change (websockets) and workload class (long-horizon tasks) where the gain is largest. Useful baseline for comparing agent runtimes.

Cited by

No citations on record.