Open-Weights Robot-Use Harness: Qwen3.8 Scores 31.39, Beating GPT-6 Astra's 28.97
aigclink · x · 2026-11-06
The paper "Democratizing Robot-Use Agents: A Unified Harness with Open Models" introduces uni-robot-agent, a unified harness for robot-use agents built on open-weight models and designed for further training. Without privileged inputs, out-of-the-box Qwen3.8-Flash-Next achieves a RoboDojo score of 31.39, beating the published GPT-6 Astra-RoboProbe result of 28.97, while staying competitive on 19 real-world tasks. The team systematically compares models across generations, families and sizes, identifying four general capabilities linked to effective robot use, and argues future training should develop them jointly. Paper, blog and code are all public.
More from Embodied
- Voice-to-app demo: PlaiPin device + Grok Gadget SDK builds apps without a keyboard — tetsuoai · 2026-11-06
- Modded MacBook Pro gets a working touchscreen — used to play Honor of Kings — xiaohu · 2026-11-06
- Roboticist shares Stanford-collab research directions and BARS talk — Fowe · 2026-11-06
- RAX From Harvard/NVIDIA/Berkeley Lifts Robot Success From 26% to 94% With Frozen Weights — jiqizhixin · 2026-11-06
- Dev builds Salvo battleship game for dual-screen iPhone Duo with head-to-head and AI play — HankYeomans · 2026-11-06
- First-hand demo of Wayve AI Driver on San Francisco streets impresses rider — Sethwinterroth · 2026-11-06