IQuest Q1 debuts: 320B/15B coding model with 512K context and hybrid attention

AdinaYakup · x · 2026-09-29

IQuest Research released IQuest Q1, a new coding model in 320B and 15B sizes under a custom license. It offers a 512K context window using hybrid attention (3 sliding-window layers per 1 full-attention layer) for efficient long-context handling, and the team is notably transparent about its limitations.

Related event: IQuest Open-Sources IQuest-Q1, a 320B MoE Model for Agentic Coding(11 posts)→

Original post →

More from coding & agent

coding & agent channel →