Preprint · 2026 · Open access

Memory That Doesn't Expire: Agentic Long-Term Memory for Conversational Agents on a Single Consumer GPU

Muhammad Abdullah Shaheer

Zenodo (CERN European Organization for Nuclear Research) · Published 16 August 2026 · DOI 10.5281/zenodo.21964737

Open full textCite this paper

In brief

“Memory That Doesn't Expire: Agentic Long-Term Memory for Conversational Agents on a Single Consumer GPU” is a 2026 preprint by Muhammad Abdullah Shaheer, published in Zenodo (CERN European Organization for Nuclear Research). OpenAlex counts 0 citations as of 19 November 2026. It is open access under a CC BY licence: the full text is free to read at Zenodo (CERN European Organization for Nuclear Research).

Citations
0
Authors
1

Abstract

Conversational agents forget: once a dialogue outgrows the context window, a model must either be reminded through retrieval or have its history truncated. Recent memory systems address this with extraction and retrieval, but they assume a large, often API-hosted model as the memory manager, an assumption that rules out private, local deployment. We ask whether a small local model, on a single consumer GPU, can manage long-term memory well, and how the answer depends on model size. We evaluate four memory strategies (no memory, full-history stuffing, raw-turn retrieval, and an agentic protocol that extracts typed, dated, deduplicated facts) on LoCoMo (1,986 questions over 10 long multi-session conversations), scored by an independent LLM judge. Full-context stuffing wins aggregate accuracy when the answer still lies inside the window, but its advantage is an artifact of conversation length: split by whether a question's evidence survives truncation, stuffing collapses by 76% beyond the window while agentic memory stays flat, outperforming stuffing 2.0x on the 39% of questions that already exceed it. Agentic memory further gives the best temporal reasoning of any strategy (via absolute-date grounding at write time) using ~67x less context. Varying the extractor from 3B to 14B, we find quality scales with size and that temporal grounding is the first capability to break under compression, setting a minimum viable memory-manager size of ~7-14B, comfortably within a 12 GB GPU. A hand-verified audit of 200 stored memories finds 199 faithful to their source turn, with no fabrications. We release the system, the benchmark harness, and the raw judged outputs behind every reported number. Code: https://github.com/MasHasNoLife/yuna-aiPaper source: https://github.com/MasHasNoLife/memory-that-doesnt-expireRaw run outputs: https://doi.org/10.5281/zenodo.21702651

Cite this paper

Formatted from the catalogue record. Check names, capitals and page numbers against the paper before you submit.

APA 7

Shaheer, M. A. (2026). Memory That Doesn't Expire: Agentic Long-Term Memory for Conversational Agents on a Single Consumer GPU. Zenodo (CERN European Organization for Nuclear Research). https://doi.org/10.5281/zenodo.21964737

MLA 9

Shaheer, Muhammad Abdullah. “Memory That Doesn't Expire: Agentic Long-Term Memory for Conversational Agents on a Single Consumer GPU.” Zenodo (CERN European Organization for Nuclear Research), 2026. https://doi.org/10.5281/zenodo.21964737.

Chicago

Shaheer, Muhammad Abdullah. 2026. “Memory That Doesn't Expire: Agentic Long-Term Memory for Conversational Agents on a Single Consumer GPU.” Zenodo (CERN European Organization for Nuclear Research). https://doi.org/10.5281/zenodo.21964737.

Harvard

Shaheer, M. A. (2026) ‘Memory That Doesn't Expire: Agentic Long-Term Memory for Conversational Agents on a Single Consumer GPU’, Zenodo (CERN European Organization for Nuclear Research). doi:10.5281/zenodo.21964737.

IEEE

M. A. Shaheer, “Memory That Doesn't Expire: Agentic Long-Term Memory for Conversational Agents on a Single Consumer GPU,” Zenodo (CERN European Organization for Nuclear Research), 2026, doi: 10.5281/zenodo.21964737.

BibTeX

@misc{shaheer2026memory,
  title     = {{Memory That Doesn't Expire: Agentic Long-Term Memory for Conversational Agents on a Single Consumer GPU}},
  author    = {Shaheer, Muhammad Abdullah},
  howpublished = {Zenodo (CERN European Organization for Nuclear Research)},
  year      = {2026},
  doi       = {10.5281/zenodo.21964737},
  url       = {https://doi.org/10.5281/zenodo.21964737}
}

Download for a reference manager: BibTeX (.bib) · RIS for Zotero, Mendeley and EndNote (.ris)

Topic and keywords

Physical Sciences › Computer Science › Artificial Intelligence › Speech and dialogue systems

  • long-term memory
  • chatbots
  • agentic memory
  • local LLM deployment
  • consumer GPUs
  • memory extraction
  • temporal grounding
  • temporal reasoning

Work with this paper

Most-cited papers in Computer Science

All highly cited Computer Science papers

Questions about this paper

How do I cite this paper in APA style?

Shaheer, M. A. (2026). Memory That Doesn't Expire: Agentic Long-Term Memory for Conversational Agents on a Single Consumer GPU. Zenodo (CERN European Organization for Nuclear Research). https://doi.org/10.5281/zenodo.21964737

How many times has this paper been cited?

0 times, according to OpenAlex on 19 November 2026. Google Scholar and Scopus count different sets of sources, so their numbers differ.

Is this paper free to read?

Yes. It is open access at Zenodo (CERN European Organization for Nuclear Research): sign in and open the full text from this page.

Who wrote this paper?

Muhammad Abdullah Shaheer.

What is the DOI of this paper?

10.5281/zenodo.21964737, permanently at https://doi.org/10.5281/zenodo.21964737. Sign in and open it from this page.

Bibliographic data from OpenAlex (CC0), last checked 19 November 2026. ResearchGuru does not host this paper; once you are signed in, the links open it at the publisher or the repository that does.