Data as of Jul 25, 2026 · Based on 2,718,867 AI responses across 9,511 prompts · See how Parse measures this
BERTweetgithub.com · Low evidenceBERTweet is a pre-trained language model specifically designed for English Tweets, trained on 850M tweets using the RoBERTa procedure. It normalizes raw tweets by converting user mentions and URLs into special tokens, and is available in base and large versions for NLP tasks.
Parse Score