IQuest Open-Sources 320B MoE Model Q1 for Agentic Coding

IQuest has open-sourced IQuest-Q1, a 320B-parameter sparse MoE model activating 15B per token, with 524K context and day-one vLLM support, targeting agentic coding, reasoning, and multi-step tool use.

2026-09-29 ~ 2026-09-29 · 4 related posts

• IQuest-Q1 open-sourced: 320B MoE with 15B active, 524K context, day-0 vLLM support — vllm_project · 2026-09-29
• IQuest-Q1: 320B MoE model generates games from one prompt and debugs RL training — 量子位 · 2026-09-29
• IQuest-Q1 Debuts on Hugging Face: 320B MoE with 15B Active Params — Round-Argument-4984 · 2026-09-29
• IQuest-Q1 Specs Leak: 524K Context, 256 Experts, Hybrid Attention — jacek2023 · 2026-09-29