# Alibaba — Qwen-RobotManip: Alignment Unlocks Scale for Robotic Manipulation Foundation Models

- Company: Alibaba (alibaba.com)
- Announced: 2026-06-16T00:00:00+00:00
- Category: new-model
- Subject: Qwen
- Models affected: Qwen-RobotManip
- Source: https://qwen.ai/blog?id=qwen-robotmanip
- Record: https://forck.live/items/4614-qwen-robotmanip-alignment-unlocks-scale-for-robotic-manipulation-foundation

Alibaba announces Qwen-RobotManip, a generalizable Vision-Language-Action (VLA) foundation model built upon Qwen-VL, which introduces a unified alignment framework across representation, motion, and behavioral dimensions for robotic manipulation. The model uses only open-source robotic manipulation datasets and human demonstration videos to construct a ~38,100 hours pretraining corpus, and demonstrates emergent generalization capabilities across various real-robot platforms and tasks.

## Evidence

Verbatim from https://qwen.ai/blog?id=qwen-robotmanip:

> Qwen-RobotManip is a generalizable Vision-Language-Action (VLA) foundation model built upon Qwen-VL. It introduces a unified alignment framework across the representation, motion, and behavioral dimensions of manipulation, making large-scale multi-source training coherent rather than conflicting. Using only open-source robotic manipulation datasets and human demonstration videos without any proprietary data collection, Qwen-RobotManip constructs a ~38,100 hours pretraining corpus and already exhibits emergent generalization capabilities.

---

Record: https://forck.live/items/4614-qwen-robotmanip-alignment-unlocks-scale-for-robotic-manipulation-foundation
Catalogue: https://forck.live/llms.txt
Feed: https://forck.live/feed.md