IQuest-Q1 debuts: 320B sparse MoE foundation model for agentic CLI coding
iQuestResearch · hf · 2026-10-23
IQuest Research released IQuest-Q1, a sparse MoE agentic foundation model with 320B total / 15B activated parameters, aimed at agentic CLI systems.
- Three-stage pipeline: stable pre-training (Stable SplitNorm, Muon, μP, coding-heavy curriculum), fine-grained mid-training with progressive context extension, and multi-scaffold post-training (SFT + domain RL + MOPD)
- Model merging combines experts and checkpoints across stages while mitigating update conflicts
- Targets strong agentic coding and general capabilities
More from coding & agent
- 35 CVEs tied to AI coding tools this year, with no review budget to match — YvesMulkers · 2026-10-23
- Indie dev's agent tracks 519 directories, 311 submitted for SEO backlinks — fekdaoui · 2026-10-23
- ScienceClaw Hackathon at MIT Media Lab to build decentralized AI agent collectives for science — ProfBuehlerMIT · 2026-10-23
- What should a native agent harness handle beyond tool calling? — Glittering812 · 2026-10-23
- Running ~100 DeepSeek v4.1 flash agents in parallel for under $20/month — zack_overflow · 2026-10-23
- MCP's stateless 2026-07-28 spec drops session-id — how are stateful tools adapting? — Stock-Pumpkin-8859 · 2026-10-23