Manycore Tech

🚀 The 3D Gaussian browser is here! Manycore Tech has officially open-sourced Aholo Viewer – a 3D Gaussian browser that works across devices. No client installation needed. Just one link, and you can explore massive 3D scenes with 1B+ Gaussian points. Today marks the internet's true 3D moment! 👨💻 Aholo Viewer is now live on GitHub. Devs worldwide, go check it out github.com/manycoretech/a… #PhysicalAI #ManycoreTech #3DGaussian #SpatialIntelligence #AI #3DReconstruction

See you in Shanghai! 🥰 We're heading to #WAIC2026 with the latest in spatial intelligence — including a major LuxReal upgrade and live 3D generation demos. 👀 🕐Jul 17–20, 2026 🚩Shanghai World Expo Exhibition & Convention Center 🎮Booth H3-B110 See you there! 🙌 #WAIC2026 #SpatialIntelligence #AI #ManyCoreTech #AIVideo #3DGeneration #DigitalTwin #Innovation

@mikeroberts3000 Thanks, Mike! The SPEAR team built something genuinely exciting — glad we could help spread the word. 🙌

@ManycoreTech I love this! Beautifully summarized 😆

@mikeroberts3000 @Adobe @intel @nvidia @ETH @imperialcollege 🥰🥰🥰

@ManycoreTech @Adobe @intel @nvidia @ETH @imperialcollege 🥳🥳🥳

SPEAR is officially live. Control any #UE5 game from Python — 14K+ functions, any entity, 1080p 73+ FPS straight to NumPy. Proud to have @ManycoreTech contribute to this alongside @Adobe, @intel, @nvidia, @ETH, and @imperialcollege. 🚀 #ECCV2026 #UnrealEngine #EmbodiedAI #PhysicalAI #SpatialIntelligence #ComputerVision #OpenSource #Simulation #Robotics #WorldModels

@mikeroberts3000 Congrats Mike! Proud to collaborate on SPEAR with you and the team. 🔥

The SPEAR project page is officially live! #ECCV2026 SPEAR lets you control any #UE5 #UnrealEngine game from Python. You can call over 14K UE functions, control any game entity, and render at 1080p 73+ FPS directly to NumPy. mikeroberts3000.github.io/papers/spear arxiv.org/abs/2607.06701

Hi Mike — thanks for the reminder! We've added a full author credit reply to the original thread. Every contributor to SPEAR deserves recognition, and we're glad to have the complete team highlighted. Appreciate your support, and congratulations again to everyone on the ECCV acceptance! 🎉

@ManycoreTech Hi @ManycoreTech, please include Matthias Mueller, Stephan Richter, and Yannick Hold-Geoffroy in future PR communications about SPEAR. They are co-authors on the SPEAR paper and each contributed a ton to the project. Thank you.

Our papers just got accepted at #ECCV2026 — and the one we're most excited about: SPEAR, our next-gen Physical AI simulation platform, built with multiple tech giants. SPEAR closes the loop from real-world space to robot training: digitize → simulate → train. Alongside Syn-GRPO and WalkerBench, this is our full-stack bet on the data, simulation, and evaluation infrastructure that Physical AI runs on. Built on OpenUSD. Designed for the age of Physical AI. Huge thanks to our SPEAR co-authors and partners: @ros_german, @StefanLeuteneg1, Kalyan Sunkavalli, Vladlen Koltun, Rushikesh Zawar, Rachith Dey-Prakash, and Quentin Leboutet. #PhysicalAI #EmbodiedAI #Robotics #Simulation #ECCV2026 #SpatialAI #OpenUSD

Full credit to the entire SPEAR team on the ECCV acceptance! 🎉 Mike Roberts, Renhan Wang, Rushikesh Zawar, Rachith Dey-Prakash, Quentin Leboutet, Stephan R. Richter, Matthias Müeller, German Ros, Rui Tang, Stefan Leutenegger, Yannick Hold-Geoffroy, Kalyan Sunkavalli, Vladlen Koltun.

Our papers just got accepted at #ECCV2026 — and the one we're most excited about: SPEAR, our next-gen Physical AI simulation platform, built with multiple tech giants. SPEAR closes the loop from real-world space to robot training: digitize → simulate → train. Alongside Syn-GRPO and WalkerBench, this is our full-stack bet on the data, simulation, and evaluation infrastructure that Physical AI runs on. Built on OpenUSD. Designed for the age of Physical AI. Huge thanks to our SPEAR co-authors and partners:@ros_german, @StefanLeuteneg1, Kalyan Sunkavalli, Vladlen Koltun, Rushikesh Zawar, Rachith Dey-Prakash, and Quentin Leboutet. #PhysicalAI #EmbodiedAI #Robotics #Simulation #ECCV2026 #SpatialAI #OpenUSD

“The next frontier for video models is RL tuning with rich explanation feedback from VLMs – turning frozen gradients into precise, unhackable token-level rewards.” — Gordon Qian, Snap #RLTuning #VideoDiffusion

“Before a robot executes an action, verify if the imagined future is geometrically consistent. A broken future means a broken action.” — Zesen Zhao, University of Michigan #GeometricReasoning #WorldActionModel

Nearly 200 builders gathered at our Spatial Intelligence & World Models event @CVPR 2026 in Denver. The community is hungry for real answers on robotics, world models, video gen, and physical AI. No settled path yet. That’s what makes it interesting. Here are 6 sentences that captured the conversation. 🧵

For years, the biggest problem with 3D Gaussian Splatting wasn't quality. It was infrastructure. Huge files. No streaming. No practical way to deploy large scenes. Now we're finally seeing: ✅ LOD systems ✅ Streaming ✅ SPZ compression ✅ Collision generation ✅ Browser deployment A billion 3D Splats in a browser is impressive. What's more important is that 3DGS is starting to become practical. Huge respect to @ManycoreTech for pushing this forward 👏 Watch the full video 👉 youtu.be/2t-PLeenqqA #3DGS #GaussianSplatting #GameDev #WebGPU

🧸 Step inside a childhood dream — rebuilt with AI. User "Haozi" used Manycore’s Aholo Spatial Intelligence Platform to turn his toy‑filled home into a full 3D room tour. Digimon, Kabuto, Buzz Lightyear, Ultraman… all the memories, now explorable. Happy Children’s Day to everyone. May you never grow up inside. 🌟 #Manycore #Aholo #ChildrensDay #3DRoomTour #SpatialIntelligence #3DGS