Sequoia: Speculative decoding boosting LLM inference by 8-10x (infini-ai-lab.github.io)

3 points by fgfm 2y ago ↗ HN

0 comments

[ 10.5 ms ] story [ 4.5 ms ] thread

No comments yet.

A showcase of shard-db. Data from the Hacker News API. Refreshed every 15 minutes.

Source · Live DB stats