Miami startup Subquadratic claims 1000x AI efficiency with subquadratic LLM researchers demand independent proof

Published on 6 May 2026

The core math claim would collapse long context costs

Miami startup Subquadratic says its SubQ 1M-Preview LLM finally escapes the quadratic attention cost that has constrained major models since 2017. It claims up to 1,000x attention-compute reductions and launches an API, coding agent, and search tool after a $29 million seed. But researchers question cherry-picked benchmarks and missing pricing, calling for independent validation.

Subquadratic claims linear scaling via fully subquadratic attention architecture
It reports massive efficiency gains, especially at 1M-token contexts
Critics flag narrow benchmarks, single-run testing, and unexplained research-to-production gaps
Prior long-context claims from similar startups make the independent-proof threshold higher

#startups #attention #ai #venture capital #llm

Read the full story at Venture Beat

This summarization was done by Beige for a story published on Venture Beat

Miami startup Subquadratic claims 1000x AI efficiency with subquadratic LLM researchers demand independent proof

The full experience is on mobile.