Request invite
Research · 23 October 2026
Research

IQuest-Q1: 320B Sparse Mixture-of-Experts Agentic Foundation Model with 15B Activated Parameters

IQuest-Q1 is a sparse MoE agentic foundation model with 320B total parameters, 15B activated, developed via a three-stage process combining Stable SplitNorm, Muon, muP, and a curriculum that increases coding-data mixture.

Why it matters

Benchmark IQuest-Q1 against existing agentic models; consider its 15B active capacity for scaling.

For product managers

For your product
What does this mean for your product?
Try it on your product
Read at Hugging Face Papers
More from this issueAll AI news from 23 Oct →
Google Cloud Launches Agent Compute Engine: A MicroVM per AI AgentAnnounced
OpenAI launches Decisions API to compete with TypeSafe's JevAvailable now
Recursive releases Flint, 40B MoE model built by AI system 'Eureka Machine'Available now
OpenAI launches Codex and models on Google Cloud's Model GardenAnnounced

Get this every morning, written for your product.

Deltaday reads the whole AI news cycle and tells you what changed, whether it's real yet, and what to do about it.

Request an inviteSee how it works