Qwen Releases Qwen3.8-Max: 2.4T Parameters with 1M Context natively
Scobleizer · x · 2026-08-13
Qwen has released its largest open-weight model to date: Qwen3.8-Max.
- Core Specs: 2.4T total parameters with 95B active, supporting a 1M token context window.
- Capabilities: Features adjustable reasoning and parallel tool calls, designed to take a goal and return finished work.
- Demonstrated Performance: Coded unattended for over 10 days, building a self-evolving harness from scratch; reproduced a research paper and beat its results in a 125-hour autonomous loop; completed a chip design from RTL to layout, cutting die area by 81%.
More from coding & agent
- DIRECT: retrieval-based next-question suggestions for conversational agents — burkov · 2026-10-16
- Claude Code and Codex now autocomplete the user's next action — tokenbender · 2026-10-16
- Hermes Agent ships official Reverse Engineer Anything plugin with one-command setup — Teknium · 2026-10-16
- MLExpert: 35B MoE Agent Trained to Autonomously Handle Machine Learning Engineering — Shangheng Du · 2026-10-16
- Cowork-Gym: Alibaba Synthesizes 64.9K Agent Work Environments, Boosting Qwen by 8.7 Points — Anni Zou · 2026-10-16
- MMSearch-OPD: Joint RL and On-Policy Distillation for Long-Horizon Multimodal Search Agents — PekingUniversity · 2026-10-16