Quail: open-source AI-SQL engine hits 1B+ input tokens/min on a single H100
sh_reya · x · 2026-09-27
Shreya, in collaboration with Modal, released Quail, an open-source AI-SQL engine that plans SQL queries and LLM inference together, reaching 1B+ input tokens per minute on a single H100 for one query.
- Core idea: reuse the DAG execution model of databases as a framework for inference, treating AI-powered data operators (like AI.join) as new primitives
- Julian Hyde (Apache Calcite) praised the old-meets-new approach, suggesting further gains from decomposing operators like AI.join into new primitives
- AI-powered data operators represent a novel inference workload class
A noteworthy open-source system for anyone optimizing LLM inference for data processing.
Related event: UC Berkeley and Modal Open-Source Quail, an AI-SQL Inference Engine(25 posts)→
More from Infra
- Gas-for-AI-datacenters forum co-located with CoRL shows energy demand spillover — carlosdponx · 2026-11-12
- a16z podcast: Mike Volpi on Cisco's 70+ acquisition playbook and the AI buildout — a16z · 2026-11-12
- AMD to pass Intel in server revenue by mid-2027 as share gains 6 points a year — BenBajarin · 2026-11-12
- Study: Reflective Data Center Roofs Actually Cool Daytime Ground by 1.8-2.2°C — SumitGup · 2026-11-12
- Tsinghua's BaGuaLu-2 tames MoE expert-load spikes at runtime, accepted to SC26 — jiqizhixin · 2026-11-12
- antirez backs low-bit DeepSeek 4.1 Flash over Qwen 27B dense on Apple M5 Macs — antirez · 2026-11-12