Ant Group Open-Sources LingBot-Video for Embodied AI
Ant Group's Robbyant has officially open-sourced LingBot-Video, a video foundation model designed for embodied AI. Rather than pursuing traditional visual aesthetics, the model focuses on understanding real physical laws, aiming to serve as a foundational component for a robotic 'brain' in applications like data engines, policy evaluation, and action planning.
Key Details and Architecture
LingBot-Video utilizes a fully customized DiT + MoE (Mixture of Experts) architecture. The model has a total parameter count of 30B but activates only 3B parameters during inference. This design significantly boosts inference efficiency over dense architectures when processing 1M long token sequences. The team claims it is about 3x faster on long inputs than dense models, making it suitable for real-time robotic planning. Additionally, the model supports interactive use, generating videos up to 1 hour long, and includes an agent to drive the process. It was pre-trained on large-scale internet video data and incorporated 70,000 hours of embodied data.
Training Mechanism and Evaluation
Changing the traditional scoring mechanism that only pursues visual beauty, the team employed a single-step GRPO algorithm with 6 precise reward signals: visual quality, text-image alignment, motion dynamics, motion coherence, human motion consistency, and physical rationality. This prevents the model from 'faking' physical effects. In the public RBench benchmark, LingBot-Video achieved an average score of 0.620, leading other open-source models in manipulation, long-horizon, and quadruped tasks, though the team admitted it still trails the closed-source Wan 2.6 in spatial and robotic arm tasks.
Open Source Availability
The model's code and weights have been fully open-sourced under the Apache 2.0 license. Official resources including model files, code repositories, research papers, and project pages have been released for developers to access and reproduce.
2026-07-09 ~ 2026-07-10 · 26 related posts
- LingBot-Video: Open-Source Video World Model — Savings-Display5123 · 2026-07-09
- LingBot-Video Open-Source Release — Different_Fix_2217 · 2026-07-09
- Open-Source Video Model LingBot-Video — Adventurous_Rush1474 · 2026-07-09
- [source] LingBot-Video Foundation Model Released — _akhaliq · 2026-07-09
- Open-Source Video Model Focuses on Physical Reasoning — dair_ai · 2026-07-09
- Open-Source Video Model Generates Up to One Hour — chongdashu · 2026-07-09
- Ant Group Open-Sources Embodied Video Model — 量子位 · 2026-07-09
- LingBot-Video is Now Open Source — 赛博禅心 · 2026-07-09
- Ant Lingbo Open-Sources Embodied Video Foundation — 机器之心 · 2026-07-09
- Open-Source Embodied Video Foundation Model — nikola_mr64990 · 2026-07-09
- LingBot-Video Performance and MoE Inference Acceleration — thetripathi58 · 2026-07-09
- [source] LingBot-Video Training: 6D Physical Rewards and Real Robot Data — thetripathi58 · 2026-07-09
- [source] Ant's LingBot-Video: Open-Source MoE Video Model for Robotics — thetripathi58 · 2026-07-09
- Ant Group Open-Sources LingBot-Video Weights and Code — thetripathi58 · 2026-07-09
- LingBot-Video Focuses on Physical Consistency — Div_pradeep · 2026-07-09
- Ant Group Open-Sources Robotics Video Model — 赛博禅心 · 2026-07-09
- LingBot-Video: Significance and Benchmark Scores — dr_cintas · 2026-07-09
- LingBot-Video Model Resources Roundup — dr_cintas · 2026-07-09
- Open-Source Embodied Video Generation Model Released — rohanpaul_ai · 2026-07-10
- LingBot-Video: Embodied Video Foundation Model — aigclink · 2026-07-10
- Open-Source Video MoE Model for Embodied AI — AdinaYakup · 2026-07-10
5 near-duplicate retellings: _akhaliq · thetripathi58 · dr_cintas · rohanpaul_ai · aigclink