Vektor has released version 2.0.2 of its memory benchmark tool, LongMemEval, demonstrating significant improvements in its ability to recall and process information from long chat histories. The updated build achieved an 89.2% score on a paired check of 120 LongMemEval questions, up from 84.2% in the previous version, and an 84.0% score on the full 500 questions. Key enhancements include improved temporal accuracy by converting relative dates to specific timestamps, and dynamic adjustment of memory recall size based on question complexity, with counting and totaling questions now benefiting from up to 80 memories instead of 12. AI
IMPACT Enhances LLM capabilities in handling long contexts, potentially improving user experience in conversational AI applications.
RANK_REASON This is a tool release and benchmark update, not a frontier model release.
AI-generated summary · Google Gemini · from 1 sources. How we write summaries →
