

Frances Marsha
11.7K posts

@Tech_Marsha
AI & Web Dev Enthusiast | Personal Branding | Ghost-writing | AI | Tech | News | CPP - @yapper_so | DM or Mail for Paid Promotion ✉️ [email protected]







🔥 New week, New 𝗦𝗲𝗻𝘀𝗲𝗡𝗼𝘃𝗮-𝗨𝟭 Drop — and this one goes Deep!🔥 📄 𝗧𝗵𝗲 𝗳𝘂𝗹𝗹 𝗧𝗲𝗰𝗵𝗻𝗶𝗰𝗮𝗹 𝗥𝗲𝗽𝗼𝗿𝘁 𝗶𝘀 𝗢𝗨𝗧 — the most detailed disclosure yet of how to build a frontier Native Multimodal Model. Inside: ✨ Near-lossless visual interface (no VEs, no VAEs) ✨ Native Multimodal Unified Modeling ✨ Joint AR + pixel-space flow matching training ✨ Native Mixture-of-Transformers backbone ✨ 6-stage training recipe + RL post-training + distillation If you work on NMM, this is the playbook. 🤗 One more thing: 𝗦𝗲𝗻𝘀𝗲𝗡𝗼𝘃𝗮-𝗨𝟭-𝗔𝟯𝗕-𝗠𝗼𝗧 (𝟯𝟴𝗕-𝗔𝟯𝗕 𝗠𝗼𝗘) 𝘄𝗲𝗶𝗴𝗵𝘁𝘀 𝗮𝗿𝗲 𝗻𝗼𝘄 𝗼𝗽𝗲𝗻-𝘀𝗼𝘂𝗿𝗰𝗲𝗱 — a RARE native unified model on an MoE backbone (Only 3B active! Lightning Fast⚡) 📄 Tech Report: arxiv.org/abs/2605.12500 🤗 Daily Papers (Vote & Discuss): huggingface.co/papers/2605.12… 🤗 Models: huggingface.co/collections/se… 💻 Code: github.com/OpenSenseNova/… 🎮 Demo: unify.light-ai.top 👾 Discord: discord.com/invite/BuTXPHm…



🔥 New week, New 𝗦𝗲𝗻𝘀𝗲𝗡𝗼𝘃𝗮-𝗨𝟭 Drop — and this one goes Deep!🔥 📄 𝗧𝗵𝗲 𝗳𝘂𝗹𝗹 𝗧𝗲𝗰𝗵𝗻𝗶𝗰𝗮𝗹 𝗥𝗲𝗽𝗼𝗿𝘁 𝗶𝘀 𝗢𝗨𝗧 — the most detailed disclosure yet of how to build a frontier Native Multimodal Model. Inside: ✨ Near-lossless visual interface (no VEs, no VAEs) ✨ Native Multimodal Unified Modeling ✨ Joint AR + pixel-space flow matching training ✨ Native Mixture-of-Transformers backbone ✨ 6-stage training recipe + RL post-training + distillation If you work on NMM, this is the playbook. 🤗 One more thing: 𝗦𝗲𝗻𝘀𝗲𝗡𝗼𝘃𝗮-𝗨𝟭-𝗔𝟯𝗕-𝗠𝗼𝗧 (𝟯𝟴𝗕-𝗔𝟯𝗕 𝗠𝗼𝗘) 𝘄𝗲𝗶𝗴𝗵𝘁𝘀 𝗮𝗿𝗲 𝗻𝗼𝘄 𝗼𝗽𝗲𝗻-𝘀𝗼𝘂𝗿𝗰𝗲𝗱 — a RARE native unified model on an MoE backbone (Only 3B active! Lightning Fast⚡) 📄 Tech Report: arxiv.org/abs/2605.12500 🤗 Daily Papers (Vote & Discuss): huggingface.co/papers/2605.12… 🤗 Models: huggingface.co/collections/se… 💻 Code: github.com/OpenSenseNova/… 🎮 Demo: unify.light-ai.top 👾 Discord: discord.com/invite/BuTXPHm…



🔥 New week, New 𝗦𝗲𝗻𝘀𝗲𝗡𝗼𝘃𝗮-𝗨𝟭 Drop — and this one goes Deep!🔥 📄 𝗧𝗵𝗲 𝗳𝘂𝗹𝗹 𝗧𝗲𝗰𝗵𝗻𝗶𝗰𝗮𝗹 𝗥𝗲𝗽𝗼𝗿𝘁 𝗶𝘀 𝗢𝗨𝗧 — the most detailed disclosure yet of how to build a frontier Native Multimodal Model. Inside: ✨ Near-lossless visual interface (no VEs, no VAEs) ✨ Native Multimodal Unified Modeling ✨ Joint AR + pixel-space flow matching training ✨ Native Mixture-of-Transformers backbone ✨ 6-stage training recipe + RL post-training + distillation If you work on NMM, this is the playbook. 🤗 One more thing: 𝗦𝗲𝗻𝘀𝗲𝗡𝗼𝘃𝗮-𝗨𝟭-𝗔𝟯𝗕-𝗠𝗼𝗧 (𝟯𝟴𝗕-𝗔𝟯𝗕 𝗠𝗼𝗘) 𝘄𝗲𝗶𝗴𝗵𝘁𝘀 𝗮𝗿𝗲 𝗻𝗼𝘄 𝗼𝗽𝗲𝗻-𝘀𝗼𝘂𝗿𝗰𝗲𝗱 — a RARE native unified model on an MoE backbone (Only 3B active! Lightning Fast⚡) 📄 Tech Report: arxiv.org/abs/2605.12500 🤗 Daily Papers (Vote & Discuss): huggingface.co/papers/2605.12… 🤗 Models: huggingface.co/collections/se… 💻 Code: github.com/OpenSenseNova/… 🎮 Demo: unify.light-ai.top 👾 Discord: discord.com/invite/BuTXPHm…
