Fully Autonomous Operation.
What's driving them? The answer starts tomorrow. 🔥
#LingBot2 #Robotics #EmbodiedAI
120
128
400
523,144
@AntGroup Affiliate | Building Practical Embodied AI 🌟 Intelligence in Action, Benefits for Everyone. #SpatialIntelligence #ActionModels #WorldModel
Joined January 2026
- Posts 54
- Following 25
- Followers 1,428
- Likes 95
27 Photos and videos
Jul 1
Thanks for sharing! LingBot-Depth 2.0 is coming next week. Stay tuned!
これすごい！👀
左がD405の生データ、右がlingbot-depth処理後。
RTX 3060で75ms程度で処理している。いろいろ試してみたい。
TensorRT実装（作業中）：github.com/Ar-Ray-code/lingb…
16
3
37
2,469
Jun 30
Two months after open-sourcing, LingBot-Map lands on GitHub Trending 🚀
Take a moment to revisit our autoregressive streaming 3D reconstruction model released this April. 🌟
#OpenSource #GitHubTrending
Apr 16
We’re open-sourcing LingBot-Map — our autoregressive model for streaming 3D reconstruction from a single RGB camera. Real-time camera pose estimation. Real-time 3D scene reconstruction. No specialized hardware required.🚀
#EmbodiedAI #Robotics #3DReconstruction #ComputerVision
8
2
14
1,019
Robbyant reposted
Jun 26
Forget lidar.
One single camera.
Runs in real time & is open source:
A streaming 3D model that reconstructs scenes live, at ~20 FPS, over long sequences.
End-to-end.
Optimization tricks, cleanup steps?
Nope.
And it beats both streaming and even some offline methods.
Perception is becoming software-first.
Closer to machines that see and understand the world as it unfolds.
Thanks for sharing, @YinghaoXu1
📍Models: huggingface.co/robbyant/ling…
Project page: technology.robbyant.com/ling…
Code: github.com/Robbyant/lingbot-…
Paper: arxiv.org/abs/2604.14141
——
Weekly robotics and AI insights.
Subscribe free: 22astronauts.com
74
571
5,050
585,348
Jun 15
Amazing! #LingBotWorld
Jun 14
New Timeline-driven LingBot streams
• Image prompt events on a 3D timeline
• Playhead crosses an event → stream transitions to that world
@reactorworld
• Scrub anywhere → instant context swap, no pause
• Freeze-frame handoff (no black flash)
• Events persist across reloads
• WASD mouse-look stays live through every swap
Author the journey. Navigate it in realtime.
#Reactor #GenerativeAI #RealtimeAI #InteractiveVideo #AIart #3D #LingBot #CreativeTech #FutureOfFilm #Omnia
2
5
398
Robbyant reposted
Real-time world models are a game changer for robotics. 🤖
FlashDreams delivers up to 3× faster inference so robots can simulate, plan, and adapt on the fly.
Jun 3
World models are moving beyond offline generation towards interactive, real-time experiences.
Introducing ⚡FlashDreams⚡: an open-source high-performance inference and serving library built for autoregressive world models:
🔥 Up to 3.10× faster LingBot-World inference
🔥 Up to 2.12× faster Self-Forcing inference
🔥 Up to 1.40× faster Wan2.1 inference
🔥 8 integrated models
🔥 Multi-GPU, streaming, low-latency serving
🔥 Agentic skills that teach you how to use it
FlashDreams is designed for a new generation of AI systems that continuously evolve over time while responding to user interactions. It powers applications across robotics, autonomous vehicle simulation, gaming, and virtual worlds.
Github: github.com/NVIDIA/flashdream…
Docs: nvidia.github.io/flashdreams
Research page: research.nvidia.com/labs/sil…
Join the #flashdreams Discord channel at discord.gg/yTdHDqFP
FlashDreams is also the runtime backbone behind NVIDIA OmniDreams (github.com/nv-tlabs/omni-dre…)
1/n
#AI #WorldModels #FastInference #PhysicalAI #OpenSource #NVIDIA
2
21
125
11,643
Jun 14
This weekend, our CEO Zhu Xing and Chief Scientist Shen Yujun presented Robbyant’s latest progress at the 8th BAAI Conference 🚀
At our core lies the PHYSICAL NATIVE vision: embodied intelligence deserves purpose-built native datasets and dedicated architectures tailored for physical interaction. 🦾
Wondering how we’re turning this vision into reality? 🗓 Mark your calendars—LingBot Full-Stack 2.0 is almost here, packed with groundbreaking upgrades.👀
#Robbyant #EmbodiedIntelligence #AIInnovation #LingBot
1
5
286
Jun 13
That's super coool! Two weeks since API launch. This is what the community is building. 🔥Try it now!
Jun 12
This is wild — can you believe Lingbot can now simulate a full GTA-style mission with an actual storyline?
Grab a pistol, hop in the car, pull off the heist, then drive back. The whole arc.
And it's only been 2 weeks since we launched the Lingbot API @reactorworld . The journey ahead are endless! 🚀
6
538
Jun 9
🚀 Excited to share that LingBot-World is ranked #2 overall and #1 among all open-source models on WBench — the new comprehensive benchmark for interactive world models by @Meituan_LongCat! 🏆
📊 Scoring 88.9 on Consistency (top across all 20 models 🥇) and 71.2 on Physical realism (#2 overall 🥈) really validates the direction we've been pushing on.
🙏 Big thanks to the WBench team for building such a thorough evaluation framework — it's exactly what the community needs to move forward together.
We're not slowing down. Expect more from us very soon 👀🔥
#WorldModel #OpenSource #WBench #LingBotWorld
7
3,315
Jun 5
Great work! Thrilled to see LingBot-World top the leaderboard.
Jun 5
World models feel like the future... almost... We can still see some weird artifacts.
That means we need high-quality benchmarks and the folks at @Meituan_LongCat know that.
WBench, a high-quality world generation benchmark for Interactive Video World Model Evaluation 😍
Benchmarking:
1. video quality
2. Consistency
3. Physics compliance
4. Whether the model generates what it's told (important)
🏆 Current top contenders
🥇 @robbyant_brain
🥈 @TencentHunyuan
🥉 @TencentHunyuan
I can't wait for models to generate real-time, super-immersive worlds.
1
4
744
Jun 4
So excited to share that LingBot-World is now supported by SGLang-Diffusion! 🎉 Huge props to the SGLang team for making this happen.
Sub-second latency, 30fps—developers, check the cookbook: docs.sglang.io/cookbook/diff… Let’s build! 💫
SGLang-Diffusion now supports LingBot World, our first realtime world model.
On H200-class GPUs, SGLang-Diffusion achives sub-second chunk latency, and with interpolation can deliver 30fps-class realtime output — among the fastest open-source paths we have measured for realtime world model serving.
👉 Take a look at the cookbook: docs.sglang.io/cookbook/diff…
@ant_oss #LingBot #inclusionAI #worldmodel
2
12
2,049
Jun 1
Thrilled to announce that we’ve officially entered a strategic partnership with @GenrobotAI ! We’re joining forces to tackle a core pain point in the embodied intelligence industry: how to scale up high-quality, physically realistic data faster. Let’s make big things happen! 💪
2
220
Robbyant reposted
Apr 16
Geometric Context Transformer for Streaming 3D Reconstruction
Contributions:
• We introduce LingBot-Map, a streaming 3D foundation model built around Geometric Context Attention (GCA), which maintains three complementary context types – anchor, pose-reference window, and trajectory memory – for efficient and consistent long-sequence streaming inference.
• We propose an efficient training recipe based on progressive training and context parallelism with a relative loss formulation for stable long-sequence optimization.
• We demonstrate that LingBot-Map achieves state-of-the-art performance on multiple benchmarks (Oxford Spires, Tanks and Temples, ETH3D, and 7-Scenes), significantly outperforming existing streaming approaches in reconstruction quality and inference speed.
6
57
440
24,718
Apr 16
We’re open-sourcing LingBot-Map — our autoregressive model for streaming 3D reconstruction from a single RGB camera. Real-time camera pose estimation. Real-time 3D scene reconstruction. No specialized hardware required.🚀
#EmbodiedAI #Robotics #3DReconstruction #ComputerVision
2
8
29
4,267
Apr 16
LingBot-Map runs at around 20 FPS, supports 10,000 frame continuous inference, and delivers state-of-the-art results across both pose estimation and reconstruction quality. On Oxford Spires: 6.42m ATE. On ETH3D: 85.70 F1. 📈
1
1
375
Apr 16
Today, LingBot-Map joins our growing embodied AI stack alongside LingBot-Depth, LingBot-VLA, LingBot-World, and LingBot-VA. Learn more:
🔗 Project page: technology.robbyant.com/ling…
💻 Code: github.com/Robbyant/lingbot-…
📄 Paper: arxiv.org/abs/2604.14141
🤗 Hugging Face: huggingface.co/robbyant/ling…
1
332
Robbyant reposted
Apr 16
🎉 After one year of teamwork, we are excited to release our 3D foundation model — LingBot-Map!
Unlike DA3/VGGT, LingBot-Map is a purely autoregressive model for streaming 3D reconstruction ⚡
It achieves ~20 FPS on 518×378 resolution over sequences exceeding 10,000 frames — and beyond 🚀
Two key insights behind LingBot-Map:
🔑 Keep SLAM's structural wisdom: build Geometric Context Attention with long-context modeling while maintaining a compact streaming state
🔑 Make everything end-to-end learnable — no optimization, no post-processing
Let's check out our demos 👇
97
490
4,682
1,367,198
Robbyant reposted
Mar 31
A new large-scale RGB-D dataset from Ant Group: LingBot-Depth 🤖 just dropped on @huggingface
huggingface.co/datasets/robb…
✨3M samples / 2.7TB
✨Real-world simulation VLA robotics data
✨Raw sensor depth ground truth
2
11
72
5,453
Mar 30
🚀 Exciting news for the spatial perception community!
📷 For too long, the lack of large-scale, real-world depth datasets has been a major bottleneck. Today, we are open-sourcing the RGB-D dataset built for training our spatial perception model LingBot-Depth — and it's massive. 👇
1
16
104
20,675
Mar 30
3 MILLION carefully annotated RGB-D pairs:
✅2M real-world captured samples
✅1M high-quality rendered samples
Each sample includes an RGB image, a raw sensor depth map, and a ground truth depth map — everything you need for depth estimation research.🔍
1
1
2
853
Mar 30
We built this dataset to push the boundaries of 3D vision, embodied AI, and robotics — and now it's yours.
Let's advance the open-source ecosystem for spatial perception together.
🔗 huggingface.co/datasets/robb…
2
4
756