@AravSrinivas 在推特宣布开源其团队自研并已部署到生产的tokenizer,声称效率远超Hugging Face和SentencePiece,并强调毫秒级延迟优化的重要性。
Every millisecond matters. We’re open sourcing the tokenizer we built and deployed on production; that’s far efficient than huggingface and sentencepiece.
likes: 128 | retweets: 6 | replies: 17 | views: 18417