EverMind(@evermind):From 8,594 tokens to 1,946 per answer. EverOS selects the memories a question actually needs instead of injecting a fixed 20 every time. In our LongMemEval-S test, that meant 77% fewer answer-stage input tokens, with 91% accuracy versus 93.4%. Less context to process. Relevant evidence to work with.

EverMind

@evermind

Long-term agent memory | self-improving | agent harness

Joined November 2025

22 Following 4.7K Followers

EverMind@evermind

2026.09.09 10:05

From 8,594 tokens to 1,946 per answer. EverOS selects the memories a question actually needs instead of injecting a fixed 20 every time. In our LongMemEval-S test, that meant 77% fewer answer-stage input tokens, with 91% accuracy versus 93.4%. Less context to process. Relevant evidence to work with.