LuthorSQL: a 7B one-pass text-to-SQL gate that cuts agent API costs 33.3%
Ibrahim Memon · hf · 2026-11-17
UC Irvine researchers present LuthorSQL, a non-autoregressive 7B text-to-SQL model acting as a fast System 1 gate for a slower LLM agent, escalating only uncertain queries. It hits 68.03% execution accuracy on BIRD's hidden test at 0.84s/question on one GPU, transfers zero-shot across five SQL dialects, and in a confidence cascade with Claude Opus 5.5 answers nearly half of queries locally on post-cutoff databases, cutting API cost 33.3% at equal accuracy. Contamination probing shows Opus reproduces 48–92% of public benchmark gold queries verbatim without schemas.
More from coding & agent
- HTTP 200 doesn't mean success: production lessons from an Azure AI translation workflow — WirelessLife · 2026-11-17
- LlamaIndex launches ParseBench-2 Metric Lab for customizable parser rankings — llama_index · 2026-11-17
- LlamaIndex's ParseBench Metric Lab ranks 19 doc parsers by your business needs — llama_index · 2026-11-17
- MIRIX-2.0 multi-agent memory system tops LoCoMo at 91.4%, beats best baseline by 14.3 points — _akhaliq · 2026-11-17
- Google's Nov 10 Core Update: early data shows YMYL site down ~60%, affiliate template pages hit — gaganghotra_ · 2026-11-17
- "No AI Slop Manifesto" passes 660 signatures from OpenAI, SpaceX builders — Madisonkanna · 2026-11-17