SenseTime (@SenseTime_AI) - Twitter Profili | Zamantika Mersobahis Locabet

Sabitlenmiş Tweet

SenseTime@SenseTime_AI·2h

📢📢 𝗜𝗻𝘁𝗿𝗼𝗱𝘂𝗰𝗶𝗻𝗴 𝗮𝗻 𝗲𝗻𝗵𝗮𝗻𝗰𝗲𝗱 𝗺𝗼𝗱𝗲𝗹 𝗳𝗼𝗿 𝗶𝗻𝗳𝗼𝗴𝗿𝗮𝗽𝗵𝗶𝗰 𝗴𝗲𝗻𝗲𝗿𝗮𝘁𝗶𝗼𝗻! 📊 Posters, charts, recipe cards, postcards — even arXiv-style pages — all from 𝗦𝗲𝗻𝘀𝗲𝗡𝗼𝘃𝗮-𝗨𝟭-𝟴𝗕-𝗠𝗼𝗧-𝗜𝗻𝗳𝗼𝗴𝗿𝗮𝗽𝗵𝗶𝗰. 🚀 +6.8 / +18.2 on BizGenEval (hard) / IGenBench (Q-ACC) over base U1, plus 100+ diverse showcases. 🤗 huggingface.co/sensenova/Sens… 🖼️Showcases: github.com/OpenSenseNova/… Try it out — we'd love to see what you build! @huggingface

English

7

3

13

365

SenseTime@SenseTime_AI·2h

ZXX

1

0

2

58

SenseTime@SenseTime_AI·2h

ZXX

3

0

4

80

SenseTime@SenseTime_AI·2h

📢📢 𝗜𝗻𝘁𝗿𝗼𝗱𝘂𝗰𝗶𝗻𝗴 𝗮𝗻 𝗲𝗻𝗵𝗮𝗻𝗰𝗲𝗱 𝗺𝗼𝗱𝗲𝗹 𝗳𝗼𝗿 𝗶𝗻𝗳𝗼𝗴𝗿𝗮𝗽𝗵𝗶𝗰 𝗴𝗲𝗻𝗲𝗿𝗮𝘁𝗶𝗼𝗻! 📊 Posters, charts, recipe cards, postcards — even arXiv-style pages — all from 𝗦𝗲𝗻𝘀𝗲𝗡𝗼𝘃𝗮-𝗨𝟭-𝟴𝗕-𝗠𝗼𝗧-𝗜𝗻𝗳𝗼𝗴𝗿𝗮𝗽𝗵𝗶𝗰. 🚀 +6.8 / +18.2 on BizGenEval (hard) / IGenBench (Q-ACC) over base U1, plus 100+ diverse showcases. 🤗 huggingface.co/sensenova/Sens… 🖼️Showcases: github.com/OpenSenseNova/… Try it out — we'd love to see what you build! @huggingface

English

7

3

13

365

SenseTime@SenseTime_AI·16h

This is the frontier our innovators strive for. Excited to see passionate minds driving it forward.🔥

Zhongang Cai@caizhongang

Excited to have contributed to the spatial intelligence capabilities of SenseNova-U1, surpassing strong baselines such as Qwen3.5 on key benchmarks including VSI-Bench. We’re also thrilled to open-source SenseNova-SI-8M, which is currently the largest spatial QA dataset to date. See you at CVPR this June, happy to chat in person! SenseNova-SI-8M: huggingface.co/datasets/sense…

English

4

0

7

383

SenseTime@SenseTime_AI·1d

Led by our Co‑Founder and Chief Scientist Dr. @lindahua , our passionate AI pioneers have brought vision to life.🔥 Access the 𝗦𝗲𝗻𝘀𝗲𝗡𝗼𝘃𝗮 𝗨𝟭 𝘁𝗲𝗰𝗵𝗻𝗶𝗰𝗮𝗹 𝗿𝗲𝗽𝗼𝗿𝘁 for the architecture, training recipe, and innovations behind this breakthrough.

Dahua Lin@lindahua

Proud to announce the release of the SenseNova U1 Tech Report — together with the a new set of model weights based on MoE. We hope this open release promotes transparency, reproducibility, and further innovation across the AI community. Huge thanks to the team for making this possible. 🚀

English

3

0

2

604

SenseTime@SenseTime_AI·1d

🚀 Huge thanks to Bijan Bowany for his strong recognition of 𝗦𝗲𝗻𝘀𝗲𝗡𝗼𝘃𝗮 𝗨𝟭’s strengths, proven through hands‑on testing on 𝗖𝗼𝗺𝗳𝘆𝗨𝗜. ✅ 𝗜𝗻𝗳𝗼𝗴𝗿𝗮𝗽𝗵𝗶𝗰: “𝘉𝘪𝘤𝘦𝘱 𝘤𝘶𝘳𝘭 𝘱𝘰𝘸𝘦𝘳𝘦𝘥 𝘤𝘰𝘮𝘱𝘶𝘵𝘦𝘳” → 𝗰𝗼𝗺𝗽𝗹𝗲𝘅 𝗶𝗱𝗲𝗮, 𝗰𝗹𝗲𝗮𝗻 𝘃𝗶𝘀𝘂𝗮𝗹𝘀 ✅ 𝗜𝗻𝘁𝗲𝗿𝗹𝗲𝗮𝘃𝗲𝗱 𝘁𝗲𝘅𝘁 𝗶𝗺𝗮𝗴𝗲: 𝘉𝘢𝘤𝘬 𝘵𝘰 𝘵𝘩𝘦 𝘍𝘶𝘵𝘶𝘳𝘦 𝘵𝘦𝘤𝘩 𝘳𝘦𝘷𝘪𝘦𝘸 → 𝗹𝗼𝗴𝗶𝗰𝗮𝗹 𝗰𝗼𝗵𝗲𝗿𝗲𝗻𝗰𝗲 + 𝘀𝗲𝗹𝗳 𝗿𝗲𝗳𝗹𝗲𝗰𝘁𝗶𝗼𝗻 ✅ 𝗖𝗼𝗺𝗶𝗰: “𝘏𝘶𝘮𝘢𝘯 𝘵𝘶𝘳𝘯𝘴 𝘪𝘯𝘵𝘰 𝘩𝘦𝘭𝘪𝘤𝘰𝘱𝘵𝘦𝘳” → 𝘄𝗶𝗹𝗱 𝗽𝗿𝗼𝗺𝗽𝘁𝘀 𝗵𝗮𝗻𝗱𝗹𝗲𝗱 𝘄𝗶𝘁𝗵 𝗲𝗮𝘀𝗲 🎥 Watch the video youtube.com/watch?v=-jFXyw… 👉 Try it: Hugging Face / GitHub

YouTube

English

5

3

11

1K

SenseTime@SenseTime_AI·1d

One model. Understanding and generating text and images together — not stitched, not patched. ✨Thank you @SarahAnnabels for this brilliant showcase. The crisp editorial layout that brings the story into focus — this is NEO‑Unify in action. 🙌

SARAH@SarahAnnabels

Instead of stitching models together, SenseNova U1 runs everything inside one unified system. Built on NEO-Unify, it treats text and images as part of the same space. No switching between components. No translation gaps. Just one model that understands and creates at the same time.

English

0

1

4

332

SenseTime retweetledi

Charles Roman@pumpkherm·2d

While everyone is scaling up VE/VAE, SenseNova U1 goes VE‑free + MoT. Full tech report now public. And SenseNova-U1-A3B-MoT (38B-A3B MoE) weights are now open-sourced.

SenseTime@SenseTime_AI

🔥 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…

English

15

5

23

4.7K

SenseTime retweetledi

Future Stacked@FutureStacked·2d

Most multimodal models still treat text and images like separate problems. One part reads. Another part reasons. Another part generates. SenseNova U1 is trying something different: one unified model where language and vision work inside the same architecture.

English

18

49

160

82.4K

SenseTime retweetledi

AK@_akhaliq·2d

SenseNova-U1 Unifying Multimodal Understanding and Generation with NEO-unify Architecture

English

1

7

32

7K

SenseTime@SenseTime_AI·2d

Thank you for introducing SenseNova U1 to the community. A little note: SenseNova U1 is the next‑generation native unified multimodal model 𝗱𝗲𝘃𝗲𝗹𝗼𝗽𝗲𝗱 𝗯𝘆 𝗦𝗲𝗻𝘀𝗲𝗧𝗶𝗺𝗲, not ByteDance. We warmly welcome the community to explore and experience this innovation, and we look forward to your valuable feedback.

English

0

1

6

166

DailyPapers@HuggingPapers·2d

ByteDance released SenseNova-U1 A native unified multimodal model that handles text, image understanding and generation in a single architecture without separate visual encoders or VAEs. It unifies understanding and generation as synergistic views of one process.

English

3

11

49

3.7K

SenseTime@SenseTime_AI·2d

🔥 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…

English

14

17

128

307.1K

SenseTime@SenseTime_AI·3d

Thank you for the warm reception from the developer community — it means a great deal to us. 👐 SenseNova U1 efficiently unifies language and vision in a single model — semantic understanding, pixel-level generation, and complex infographic creation. Excited to see what you build! @HuggingModels

Hugging Models@HuggingModels

Imagine a single AI that can read text, generate images, edit photos, and even handle interleaved text+image tasks. SenseNova-U1-8B-MoT is that model: an any-to-any transformer that redefines multimodal AI. It's like having a creative studio in one model.

English

0

2

438

SenseTime@SenseTime_AI·3d

This is why we built SenseNova U1. ✨ Thank you @feesyiam for using it to spotlight what matters for child welfare. Visuals make hard conversations more accessible — and that's when AI truly matters. Keep creating. 🥰

Nafees@feesyiam

I gave it a topic. It came back with a full magazine-style infographic. Charts. Layout. Icons. Colour coding. Dense structured copy. That model is SenseNova U1. And it's open-source. 🧵

English

1

0

2

264

SenseTime retweetledi

apolinario 🌐@multimodalart·7 May

open source nano-banana now on @huggingface? SenseNova U1 is an open source unified image-text understanding & generation model (like Nano Banana/GPT-Image) by @SenseTime_AI 8B parameter unified generation model, of course, I built a Spaces demo for it 👇

English

4

7

62

4.9K

SenseTime

Keşfet