# SLICE-Mem cuts LLM memory costs by 99.5% in new framework

Digest AI · Research · published 2026-10-27T04:00:00Z

Canonical: https://digestai.news/story/slice-mem-cuts-llm-memory-costs-by-99-5-in-new-framework

## Summary

Researchers introduced SLICE-Mem, a lightweight memory framework for LLM agents that reduces reliance on generative models for memory tasks. The system uses salience-guided admission and predicate-centered extraction to filter and organize user interactions without real-time LLM calls. Instead, it defers consolidation to an offline stage, cutting token consumption by **92.4%** on Memora Weekly and **99.5%** on LongMemEval-S compared to the baseline MemOS.

Evaluated on subsets of Memora and LongMemEval-S, SLICE-Mem achieved **46.60** FAMA on Memora Weekly and **56.8%** accuracy on LongMemEval-S using **8B** and **30B** Qwen backbones. While slightly trailing MemOS’s **46.68** and **59.3%**, the framework demonstrates efficiency by minimizing LLM inference overhead while maintaining coherence in long-term agent interactions.

## Key points

- SLICE-Mem uses salience-guided admission and predicate extraction to filter user turns without real-time LLM calls
- Reduces token consumption by **92.4%** on Memora Weekly and **99.5%** on LongMemEval-S vs. MemOS baseline
- Achieved **46.60** FAMA on Memora Weekly and **56.8%** accuracy on LongMemEval-S with Qwen backbones

## Why it matters

SLICE-Mem addresses a key bottleneck in LLM agents: high inference costs for memory management. By cutting token usage by nearly 100% in some cases, it could enable cheaper, scalable long-term memory for agents without sacrificing accuracy.

## Sources

1. [SLICE-Mem: Salience-Gated Lightweight Information Curation for Efficient LLM Agent Memory](https://arxiv.org/abs/2610.33015) (arXiv cs.CL, 2026-10-27, primary source)

Part of the developing story: [The Rise of External Memory in AI Agents](https://digestai.news/thread/tencent-releases-hy-memory-30b-a3b-for-external-memory-frameworks) (3 stories)

## Cite

Digest AI, "SLICE-Mem cuts LLM memory costs by 99.5% in new framework", 27 October 2026, https://digestai.news/story/slice-mem-cuts-llm-memory-costs-by-99-5-in-new-framework

---

Written by Digest AI's editorial model from the linked sources; the sources are the record. Headlines, digests and key points are written by Digest AI and may be quoted with a link to the story page. Linked articles belong to their publishers. Terms: https://digestai.news/terms#reuse
JSON: https://digestai.news/story/slice-mem-cuts-llm-memory-costs-by-99-5-in-new-framework.json