Metraipolis › archive › dc_3fad1abcae1f

What Stops a Small Language Model From Driving a Database Agent — arXiv:2609.21341

Source
arXiv (en)
Original
https://arxiv.org/abs/2609.21341
License (as archived)
CC0-1.0
Archived
2026-10-10T02:04:09.752Z
SHA-256 of the archived text
50ff307be98513bb23c2a01c8915f7f072d3a68dce4c2517e2291b147026e86f
Stable URL of this page
https://metraipolis.com/documento/dc_3fad1abcae1f

How to cite

What Stops a Small Language Model From Driving a Database Agent — arXiv:2609.21341 — arXiv (en), https://arxiv.org/abs/2609.21341 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_3fad1abcae1f (Metraipolis archive dc_3fad1abcae1f, sha256 50ff307be98513bb23c2a01c8915f7f072d3a68dce4c2517e2291b147026e86f)
BibTeX
@misc{dc3fad1abcae1f,
  title = {What Stops a Small Language Model From Driving a Database Agent — arXiv:2609.21341},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2609.21341},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_3fad1abcae1f; sha256 50ff307be98513bb23c2a01c8915f7f072d3a68dce4c2517e2291b147026e86f},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.21341. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Small open-weight language models are assumed to fail at agentic database work because they lack the reasoning capacity for it. We test that against a production system. Over eleven days we drove the agent mode of an open-source SQL client with 39 open-weight models served locally and one hosted control, across six task surfaces: 8,199 runs, 110,711 ledger events, 14,008 refused tool calls. Of the 2,100 model-attributed agent-mode losses, 1,590, or 75.7%, came from runs that had invoked at least one tool. That majority is what survives resampling models rather than runs: it holds in 99.7% of clustered resamples and in 15 of the 22 models with at least twenty losses. Within it, transport, a run that used the tools and never got a deliverable through, is the largest class at 36.2% and capability, a run that invoked no tool at all, the smallest at 17.3%; we report that ordering as a property of this corpus rather than a general finding, since clustered by model it holds in only 74.5% of resamples. Transport failures decompose into a few mechanical argument shapes. Production ledgers record refusal codes and never the model's arguments, so these were invisible for ten days; capturing them exposed five server defects, one of which demanded a field on one tool, forbade it on the sibling that composed it, then failed the run for its absence. Five server changes, touching no model, prompt or sampling setting, moved six models by 6 to 21 cells out of 30. We also report a confound we believe affects published local-model benchmarks, ours included: with no context cap, one 7.1 GB model was admitted at its full 262,144-token window and held 51 GB on a 64 GB machine, producing runs indistinguishable in any ordinary log from a model timing out. The corpus, the scorer and a verifier that regenerates every figure are released.

Tema degli esploratori: calcolo e infrastrutture (retro-catalogo arXiv: annunci del 2026-09-21 in cs.SE).
Categorie arXiv: Software Engineering (cs.SE) ; Databases (cs.DB).
Pagina del lavoro: https://arxiv.org/abs/2609.21341 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito).
Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina.
Letto il 2026-10-10T01:43:39.832Z dalla pagina https://arxiv.org/catchup/cs.SE/2026-09-21?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).