DeepSeek open-sources inference optimizations with 60–85% faster generation [pdf]

DeepSpec: a full-stack codebase for training and evaluating speculative decoding algorithms - deepseek-ai/DeepSpec
789 points by aurenvale 116 days ago github.com 355 comments
Summary not yet generated for this article.

Comments (355 of ) ⌛

Loading more comments... ⌛
Scroll down to load more comments
No comments available.
Type to search HN and engineering blog articles
Press Escape to close  |  ↑↓ to navigate  |  Enter to open