#latency optimization

9 stories · 0 in 7 days · 0 in 30 days
Stories, products and papers tagged "latency optimization", newest first.
Sep 22
Sep 11
Jul 7
Jun 13
Jun 4
Jun 3
May 28
ModelsUS vs ChinaOfficialSelected72°17:17
Chinese contentPerplexity AI 开源 Unigram 分词器,p50 延迟比 Hugging Face 低 5 倍

做搜索或 RAG 系统的团队终于有了更快的分词方案——Perplexity 开源的这个 Unigram 分词器直接降低 5 倍延迟和 6 倍 CPU 消耗,建议有高吞吐需求的开发者立刻试一下。

Primary sourcemarktechpost@Asif RazzaqSource
May 22
May 21