SenseTime Group

02/09/2026

🚀 𝗦𝗲𝗻𝘀𝗲𝗡𝗼𝘃𝗮 𝗨𝟭.𝟱 𝗟𝗶𝘁𝗲 𝗔𝗜 𝗚𝗲𝗻𝗲𝗿𝗮𝘁𝗶𝗼𝗻 𝗖𝗵𝗮𝗹𝗹𝗲𝗻𝗴𝗲 𝗶𝘀 𝗟𝗜𝗩𝗘! 🎨
Create, share, and win 𝘂𝗽 𝘁𝗼 $𝟭𝟬𝟬 𝗶𝗻 𝗚𝗶𝗳𝘁 𝗖𝗮𝗿𝗱𝘀 + 𝗟𝗶𝗳𝗲𝘁𝗶𝗺𝗲 𝗨𝟭 𝗣𝗿𝗼 𝗕𝗲𝘁𝗮 𝗔𝗰𝗰𝗲𝘀𝘀!

𝗛𝗼𝘄 𝘁𝗼 𝗷𝗼𝗶𝗻:
1️⃣ Create with SenseNova U1.5 Lite via 𝗦𝘁𝘂𝗱𝗶𝗼, 𝗟𝗼𝗰𝗮𝗹, 𝗼𝗿 𝗔𝗣𝗜
2️⃣ Submit artwork to 𝗗𝗶𝘀𝗰𝗼𝗿𝗱 or 𝗦𝘁𝘂𝗱𝗶𝗼
3️⃣ Share your creation on 𝘀𝗼𝗰𝗶𝗮𝗹 𝗺𝗲𝗱𝗶𝗮 with #𝗦𝗲𝗻𝘀𝗲𝗡𝗼𝘃𝗮𝗨𝟭

🏆 𝟰 𝗧𝗿𝗮𝗰𝗸𝘀
🔹5-Day Check-in
🔹Gallery Top Voted
🔹Prompt Review
🔹Social Views (1K–3K+)

⏰ 𝗗𝗲𝗮𝗱𝗹𝗶𝗻𝗲: 𝗦𝗲𝗽 𝟭𝟬, 𝟮𝟰:𝟬𝟬 𝗖𝗦𝗧
👉 𝗝𝗼𝗶𝗻 𝗗𝗶𝘀𝗰𝗼𝗿𝗱 𝘁𝗼 𝗰𝗹𝗮𝗶𝗺 𝗿𝗲𝘄𝗮𝗿𝗱𝘀:
https://discord.gg/BuTXPHmQub

28/08/2026

We’re delighted to have participated in the Federation of Hong Kong Industries 香港工業總會 X 香港科技園 "𝗜𝗻𝗻𝗼𝗽𝗿𝗲𝗻𝗲𝘂𝗿 𝗘𝘅𝗽𝗲𝗿𝗶𝗲𝗻𝗰𝗲 𝗝𝗼𝘂𝗿𝗻𝗲𝘆" Cohort 5 (2026) this August.

Throughout the programme, secondary school students gained first-hand 𝗲𝘅𝗽𝗼𝘀𝘂𝗿𝗲 𝘁𝗼 𝘁𝗵𝗲 𝘄𝗼𝗿𝗸𝗽𝗹𝗮𝗰𝗲 𝗮𝗰𝗿𝗼𝘀𝘀 𝗱𝗶𝗳𝗳𝗲𝗿𝗲𝗻𝘁 𝗯𝘂𝘀𝗶𝗻𝗲𝘀𝘀 𝗳𝘂𝗻𝗰𝘁𝗶𝗼𝗻𝘀, 𝘄𝗶𝘁𝗵 𝘁𝗵𝗲 𝗴𝘂𝗶𝗱𝗮𝗻𝗰𝗲 𝗮𝗻𝗱 𝘀𝘂𝗽𝗽𝗼𝗿𝘁 𝗼𝗳 𝗼𝘂𝗿 𝘁𝗲𝗮𝗺𝘀. Through sharing sessions, interactive exchanges, and hands-on experience, they deepened their understanding of 𝗦𝗲𝗻𝘀𝗲𝗧𝗶𝗺𝗲’𝘀 𝗱𝗮𝘆-𝘁𝗼-𝗱𝗮𝘆 𝗼𝗽𝗲𝗿𝗮𝘁𝗶𝗼𝗻𝘀, the 𝗔𝗜 𝗮𝗻𝗱 𝗶𝗻𝗻𝗼𝘃𝗮𝘁𝗶𝗼𝗻 𝗶𝗻𝗱𝘂𝘀𝘁𝗿𝘆, 𝘁𝗲𝗮𝗺 𝗰𝗼𝗹𝗹𝗮𝗯𝗼𝗿𝗮𝘁𝗶𝗼𝗻, and 𝗱𝗶𝗳𝗳𝗲𝗿𝗲𝗻𝘁 𝗿𝗼𝗹𝗲𝘀 𝗶𝗻 𝗮 𝗿𝗲𝗮𝗹 𝗯𝘂𝘀𝗶𝗻𝗲𝘀𝘀 𝗲𝗻𝘃𝗶𝗿𝗼𝗻𝗺𝗲𝗻𝘁.

During the 𝗶𝗻𝗱𝘂𝘀𝘁𝗿𝘆 𝗲𝘅𝗰𝗵𝗮𝗻𝗴𝗲 𝗽𝗮𝗻𝗲𝗹, our expert also shared his insights on 𝗵𝗼𝘄 𝗔𝗜 𝗰𝗮𝗻 𝗰𝗿𝗲𝗮𝘁𝗲 𝗻𝗲𝘄 𝗼𝗽𝗽𝗼𝗿𝘁𝘂𝗻𝗶𝘁𝗶𝗲𝘀 𝗳𝗼𝗿 𝗛𝗼𝗻𝗴 𝗞𝗼𝗻𝗴’𝘀 𝗳𝘂𝘁𝘂𝗿𝗲 𝗱𝗲𝘃𝗲𝗹𝗼𝗽𝗺𝗲𝗻𝘁, alongside other industry representatives.

In return, our teams were inspired by the students’ fresh perspectives, thoughtful questions, and enthusiasm, making it a valuable two-way learning experience.

We were especially pleased to see our participating student Zekai Ma win Third Place in the 𝗦𝘁𝘂𝗱𝗲𝗻𝘁 𝗦𝘁𝗮𝗿 𝗔𝘄𝗮𝗿𝗱 as part of the programme. The award recognised students for their insightful reflections and key takeaways from the three-day workplace experience. Congratulations on this outstanding achievement!

A big thank you to FHKI, HKSTP, all participating students, and our colleagues supporting this meaningful journey. We look forward to continuing to nurture young talent and contributing to Hong Kong’s innovation and technology ecosystem.

28/08/2026

📈 𝗦𝗲𝗻𝘀𝗲𝗧𝗶𝗺𝗲 𝗥𝗲𝗰𝗼𝗿𝗱𝘀 𝗙𝗶𝗿𝘀𝘁-𝗘𝘃𝗲𝗿 𝗣𝗿𝗼𝗳𝗶𝘁 𝗶𝗻 𝟭𝗛 𝟮𝟬𝟮𝟲
We are pleased to report our first-ever IFRS profit since listing, marking a major milestone in our journey from AI innovation to high-quality commercialization.

🔸 𝗥𝗠𝗕 𝟮.𝟵𝟭𝗕 in 𝘁𝗼𝘁𝗮𝗹 𝗿𝗲𝘃𝗲𝗻𝘂𝗲, +𝟮𝟯.𝟰% 𝗬𝗼𝗬
🔸 𝗥𝗠𝗕 𝟭.𝟮𝟭𝗕 in 𝗴𝗿𝗼𝘀𝘀 𝗽𝗿𝗼𝗳𝗶𝘁, +𝟯𝟮.𝟵% 𝗬𝗼𝗬
🔸 𝗥𝗠𝗕 𝟲𝟮𝟬𝗠 in 𝗜𝗙𝗥𝗦 𝗻𝗲𝘁 𝗽𝗿𝗼𝗳𝗶𝘁
🔸 𝗥𝗠𝗕 𝟮.𝟯𝟯𝗕 in 𝗚𝗲𝗻𝗲𝗿𝗮𝘁𝗶𝘃𝗲 𝗔𝗜 𝗿𝗲𝘃𝗲𝗻𝘂𝗲, +𝟮𝟴.𝟮% 𝗬𝗼𝗬, accounting for nearly 80% of total revenue
🔸 𝗥𝗠𝗕 𝟭.𝟭𝟰𝗕 in 𝗿𝗲𝗰𝘂𝗿𝗿𝗶𝗻𝗴 𝗿𝗲𝘃𝗲𝗻𝘂𝗲, +𝟭𝟮𝟰.𝟰% 𝗬𝗼𝗬, accounting for 39.3% of total revenue
🔸 𝟭𝟮𝟳% 𝗬𝗼𝗬 𝗴𝗿𝗼𝘄𝘁𝗵 in 𝗼𝘃𝗲𝗿𝘀𝗲𝗮𝘀 𝗯𝘂𝘀𝗶𝗻𝗲𝘀𝘀 𝗿𝗲𝘃𝗲𝗻𝘂𝗲
🔸 𝟲𝟳.𝟯% 𝗬𝗼𝗬 𝗿𝗲𝗱𝘂𝗰𝘁𝗶𝗼𝗻 in 𝗡𝗼𝗻-𝗜𝗙𝗥𝗦 𝗮𝗱𝗷𝘂𝘀𝘁𝗲𝗱 net loss, with a substantial reduction in core business losses
🔸 𝗔𝘃𝗲𝗿𝗮𝗴𝗲 𝗱𝗮𝗶𝗹𝘆 𝗧𝗼𝗸𝗲𝗻 𝘀𝗲𝗿𝘃𝗶𝗰𝗲 𝘃𝗼𝗹𝘂𝗺𝗲 increased ~𝟮𝟮-𝗳𝗼𝗹𝗱 𝗬𝗼𝗬 in July 2026

This progress is underpinned by our Three-Pillar Framework:
🔸𝗢𝗻𝗲 𝗠𝗼𝗱𝗲𝗹 𝗦𝘆𝘀𝘁𝗲𝗺 continuously pushes the boundaries of intelligence;
🔸𝗢𝗻𝗲 𝗧𝗼𝗸𝗲𝗻 𝗙𝗮𝗰𝘁𝗼𝗿𝘆 leverages infrastructure and optimization to reduce Token costs while strengthening cost-based pricing power;
🔸𝗢𝗻𝗲 𝗔𝗴𝗲𝗻𝘁 𝗛𝗮𝗿𝗻𝗲𝘀𝘀 deeply integrates agents into enterprise and individual workflows, enabling end-to-end task ex*****on.

Together, these three pillars allow us to move beyond pure price competition by 𝗱𝗶𝗳𝗳𝗲𝗿𝗲𝗻𝘁𝗶𝗮𝘁𝗶𝗻𝗴 𝘁𝗵𝗿𝗼𝘂𝗴𝗵 𝘁𝗲𝗰𝗵𝗻𝗼𝗹𝗼𝗴𝘆 𝗮𝗻𝗱 𝘀𝗲𝗿𝘃𝗶𝗰𝗲𝘀, 𝗲𝘅𝗽𝗮𝗻𝗱𝗶𝗻𝗴 𝗼𝘂𝗿 𝘂𝘀𝗲𝗿 𝗯𝗮𝘀𝗲, 𝗮𝗻𝗱 𝗱𝗿𝗶𝘃𝗶𝗻𝗴 𝘀𝘂𝘀𝘁𝗮𝗶𝗻𝗮𝗯𝗹𝗲 𝗴𝗿𝗼𝘄𝘁𝗵 𝗶𝗻 𝗿𝗲𝗰𝘂𝗿𝗿𝗶𝗻𝗴 𝗿𝗲𝘃𝗲𝗻𝘂𝗲.

Looking ahead, we will continue to 𝗮𝗱𝘃𝗮𝗻𝗰𝗲 𝗺𝘂𝗹𝘁𝗶𝗺𝗼𝗱𝗮𝗹 𝗮𝗴𝗲𝗻𝘁𝘀, evolve the Token Factory 𝗳𝗿𝗼𝗺 𝘁𝗼𝗸𝗲𝗻-𝘃𝗼𝗹𝘂𝗺𝗲-𝗯𝗮𝘀𝗲𝗱 𝗽𝗿𝗶𝗰𝗶𝗻𝗴 𝘁𝗼𝘄𝗮𝗿𝗱 𝗼𝘂𝘁𝗰𝗼𝗺𝗲-𝗯𝗮𝘀𝗲𝗱 𝗽𝗿𝗶𝗰𝗶𝗻𝗴, and 𝗱𝗲𝗲𝗽𝗲𝗻 𝘁𝗵𝗲 𝗮𝗽𝗽𝗹𝗶𝗰𝗮𝘁𝗶𝗼𝗻 𝗼𝗳 𝗺𝘂𝗹𝘁𝗶-𝗮𝗴𝗲𝗻𝘁 𝗰𝗼𝗹𝗹𝗮𝗯𝗼𝗿𝗮𝘁𝗶𝗼𝗻 across enterprise, individual, and OPC workflows.

25/08/2026

𝗦𝗲𝗻𝘀𝗲𝗡𝗼𝘃𝗮 𝗨𝟭.𝟱 𝗟𝗶𝘁𝗲 — 𝗼𝗽𝗲𝗻-𝘀𝗼𝘂𝗿𝗰𝗲𝗱 𝗮𝗻𝗱 𝗻𝗼𝘄 𝗹𝗶𝘃𝗲 𝗼𝗻 𝘁𝗵𝗲 𝗧𝗼𝗸𝗲𝗻 𝗣𝗹𝗮𝗻: 𝗙𝗥𝗘𝗘 𝗽𝘂𝗯𝗹𝗶𝗰 𝗯𝗲𝘁𝗮, with 1,500 requests every 5 hours.

Supports integrated image generation and editing, with:
🔹 𝗥𝗲𝗳𝗲𝗿𝗲𝗻𝗰𝗲 𝗶𝗺𝗮𝗴𝗲 𝗱𝗿𝗶𝘃𝗲𝗻 creation and editing
🔹 𝗖𝗼𝗺𝗽𝗹𝗲𝘅 𝗶𝗻𝘀𝘁𝗿𝘂𝗰𝘁𝗶𝗼𝗻 𝗳𝗼𝗹𝗹𝗼𝘄𝗶𝗻𝗴
🔹 𝗡𝗮𝘁𝗶𝘃𝗲 𝟮𝗞/𝟰𝗞 𝗼𝘂𝘁𝗽𝘂𝘁
🔹 𝗔𝘂𝘁𝗼𝗺𝗮𝘁𝗶𝗰 𝗽𝗿𝗼𝗺𝗽𝘁 𝗼𝗽𝘁𝗶𝗺𝗶𝘇𝗮𝘁𝗶𝗼𝗻

🔑Get your API key: https://platform.sensenova.ai
🧩Model ID: sensenova-u1.5-lite

20/08/2026

𝗦𝗲𝗻𝘀𝗲𝗡𝗼𝘃𝗮 𝗨𝟭.𝟱 𝗟𝗶𝘁𝗲 — 𝗦𝗵𝗮𝗿𝗽𝗲𝗿. 𝗠𝗼𝗿𝗲 𝗖𝗼𝗻𝘁𝗿𝗼𝗹𝗹𝗮𝗯𝗹𝗲. 𝗠𝗼𝗿𝗲 𝗖𝗼𝘀𝘁-𝗘𝗳𝗳𝗶𝗰𝗶𝗲𝗻𝘁. An 𝗼𝗽𝗲𝗻-𝘀𝗼𝘂𝗿𝗰𝗲, lightweight native unified multimodal model for visual understanding, generation & editing.

𝗕𝘂𝗶𝗹𝘁 𝗳𝗼𝗿 𝗿𝗲𝗮𝗹-𝘄𝗼𝗿𝗹𝗱 𝘃𝗶𝘀𝘂𝗮𝗹 𝗮𝗿𝘁𝗶𝗳𝗮𝗰𝘁 𝘄𝗼𝗿𝗸𝗳𝗹𝗼𝘄𝘀:
🔹𝗡𝗮𝘁𝗶𝘃𝗲 𝟰𝗞 𝗴𝗲𝗻𝗲𝗿𝗮𝘁𝗶𝗼𝗻 with better detail, composition & realism
🔹𝗖𝗼𝗺𝗽𝗹𝗲𝘅 𝗶𝗻𝘀𝘁𝗿𝘂𝗰𝘁𝗶𝗼𝗻 𝗳𝗼𝗹𝗹𝗼𝘄𝗶𝗻𝗴 across subjects, counts, text, layouts & styles
🔹𝗘𝗻𝗵𝗮𝗻𝗰𝗲𝗱 𝗖𝗵𝗶𝗻𝗲𝘀𝗲/𝗘𝗻𝗴𝗹𝗶𝘀𝗵 𝘁𝗲𝘅𝘁 𝗿𝗲𝗻𝗱𝗲𝗿𝗶𝗻𝗴 & 𝗺𝘂𝗹𝘁𝗶-𝘁𝗲𝘅𝘁 𝗹𝗮𝘆𝗼𝘂𝘁
🔹𝗣𝗿𝗲𝗰𝗶𝘀𝗲 𝗶𝗺𝗮𝗴𝗲 𝗲𝗱𝗶𝘁𝗶𝗻𝗴 & 𝗰𝗼𝗻𝘁𝗿𝗼𝗹 with visual markers, bounding boxes & multi-image references

Just 𝟴𝗕 parameters — outperforming same-size models in instruction following and editing preservation, while 𝗰𝗼𝗺𝗽𝗲𝘁𝗶𝗻𝗴 𝘄𝗶𝘁𝗵 𝗹𝗮𝗿𝗴𝗲 𝗰𝗼𝗺𝗺𝗲𝗿𝗰𝗶𝗮𝗹 𝗺𝗼𝗱𝗲𝗹𝘀 𝗶𝗻 𝘁𝗲𝘅𝘁 𝗿𝗲𝗻𝗱𝗲𝗿𝗶𝗻𝗴 𝗮𝗻𝗱 𝗰𝗼𝗺𝗽𝗹𝗲𝘅 𝗹𝗮𝘆𝗼𝘂𝘁𝘀.

🛠️GitHub: https://github.com/OpenSenseNova/SenseNova-U1
🤗HF: https://huggingface.co/collections/sensenova/sensenova-u15
👾Discord: https://discord.com/invite/BuTXPHmQub
🎨SenseNova Studio: https://unify.light-ai.top/

19/08/2026

Overseas Business Momentum: SenseTime MEA, our joint venture with Saudi Arabia's Public Investment Fund, has 𝗽𝗮𝗿𝘁𝗻𝗲𝗿𝗲𝗱 𝘄𝗶𝘁𝗵 𝗦𝗮𝘂𝗱𝗶 𝗔𝗿𝗮𝗯𝗶𝗮’𝘀 𝗻𝗮𝘁𝗶𝗼𝗻𝗮𝗹 𝗮𝘂𝘁𝗵𝗼𝗿𝗶𝘁𝘆 𝘁𝗼 𝗱𝗲𝘃𝗲𝗹𝗼𝗽 𝗮 𝗻𝗮𝘁𝗶𝗼𝗻𝗮𝗹 𝗔𝗜-𝗽𝗼𝘄𝗲𝗿𝗲𝗱 𝗱𝗶𝗴𝗶𝘁𝗮𝗹 𝗽𝗹𝗮𝘁𝗳𝗼𝗿𝗺 𝗳𝗼𝗿 𝘀𝗰𝗵𝗼𝗼𝗹 𝘁𝗿𝗮𝗻𝘀𝗽𝗼𝗿𝘁𝗮𝘁𝗶𝗼𝗻.

Serving over 1.2 million primary and secondary students nationwide, the platform delivers 𝗶𝗻𝘁𝗲𝗹𝗹𝗶𝗴𝗲𝗻𝘁 𝘀𝗰𝗵𝗲𝗱𝘂𝗹𝗶𝗻𝗴, 𝗲𝗻𝗵𝗮𝗻𝗰𝗲𝗱 𝘀𝗮𝗳𝗲𝘁𝘆 𝗺𝗼𝗻𝗶𝘁𝗼𝗿𝗶𝗻𝗴, 𝗺𝘂𝗹𝘁𝗶-𝗽𝗮𝗿𝘁𝘆 𝗰𝗼𝗼𝗿𝗱𝗶𝗻𝗮𝘁𝗶𝗼𝗻, and 𝗱𝗮𝘁𝗮-𝗱𝗿𝗶𝘃𝗲𝗻 𝗱𝗲𝗰𝗶𝘀𝗶𝗼𝗻-𝗺𝗮𝗸𝗶𝗻𝗴 across the student transport ecosystem. Built on a scalable, secure, and resilient architecture, it provides a 𝘀𝘁𝗿𝗼𝗻𝗴 𝗳𝗼𝘂𝗻𝗱𝗮𝘁𝗶𝗼𝗻 𝗳𝗼𝗿 𝗳𝘂𝘁𝘂𝗿𝗲 𝗲𝘅𝗽𝗮𝗻𝘀𝗶𝗼𝗻 𝗮𝗻𝗱 𝗶𝗻𝗻𝗼𝘃𝗮𝘁𝗶𝗼𝗻.

We are committed to advancing the digital transformation of education and public services in Saudi Arabia, in line with the strategic objectives of 𝗦𝗮𝘂𝗱𝗶 𝗩𝗶𝘀𝗶𝗼𝗻 𝟮𝟬𝟯𝟬.

14/08/2026

Check out YouTuber NFT TIGERS SPOTON showcasing our open-source 𝗦𝗲𝗻𝘀𝗲𝗡𝗼𝘃𝗮 𝗨𝟭 with comic panel and graffiti banner generation!

🔹 Accurate text rendering
🔹 Complex, stable layout generation
🔹 Native interleaved reasoning

Try it now!

✏️SenseNova Studio: https://unify.light-ai.top/ (Try infographics; also join Discord for text-image interleaved gen)
🤗HF: https://huggingface.co/sensenova/SenseNova-U1-8B-MoT-Infographic-V2
🛠️GitHub: https://github.com/OpenSenseNova/SenseNova-U1
👾Discord: https://discord.com/invite/BuTXPHmQub
⚡️Token Plan: https://www.sensenova.ai/token-plan

📹 https://www.youtube.com/watch?v=FfoR81pQIls&t=27s

Discover Neo Unify, an open-source multimodal AI architecture that ...

13/08/2026

Thanks YouTuber ZAFFAI for highlighting our open-source 𝗦𝗲𝗻𝘀𝗲𝗡𝗼𝘃𝗮 𝗨𝟭, powering structured visual communication with sharp dense text rendering.

Featured Showcase:
📊 KPI Dashboards | 🛣️ Technical Roadmaps | 📄 Academic Summaries | 💡 Interleaved Tutorials

✏️SenseNova Studio: https://unify.light-ai.top/ (Try infographics; also join Discord for text-image interleaved gen)
🤗HF: https://huggingface.co/sensenova/SenseNova-U1-8B-MoT-Infographic-V3
🛠️GitHub: https://github.com/OpenSenseNova/SenseNova-U1
👾Discord: https://discord.com/invite/BuTXPHmQub

📹 https://www.youtube.com/watch?v=6qN4k-zHls0

🚀 Try SenseNova U1 Playground here:https://unify.light-ai.top/🤗 ...

04/08/2026

Thanks Dan for testing SenseNova U1 — our open-source AI model that reasons and generates images in one unified flow.

✅ Infographic mode: turns 1 prompt into a structured slide (icons, sections, layout)
✅ Interleaved mode: builds text + images step-by-step — watch it teach a 6-step dragon drawing tutorial

✏️SenseNova Studio: https://unify.light-ai.top/ (Try infographics; also join Discord for text-image interleaved gen)
🤗HF: https://huggingface.co/sensenova/SenseNova-U1-8B-MoT-Infographic-V2
🛠️https://github.com/OpenSenseNova/SenseNova-U1
👾Discord: https://discord.com/invite/BuTXPHmQub

📹 https://www.youtube.com/watch?v=CRxmuQPCbOo

🔥 Modelo en Hugging Face — SenseNova-U1-8B-MoT-Infographic-V2: http...

03/08/2026

𝗜𝗻𝘁𝗿𝗼𝗱𝘂𝗰𝗶𝗻𝗴 𝗦𝗲𝗻𝘀𝗲𝗡𝗼𝘃𝗮 𝗨1.5-𝗟𝗶𝘁𝗲-𝗣𝗿𝗲𝘃𝗶𝗲𝘄.

An early open-source preview of our lightweight, natively unified multimodal model — built on the native NEO-Unify architecture to natively understand, reason, generate, and edit across modalities. 𝗪𝗶𝘁𝗵 𝗷𝘂𝘀𝘁 8𝗕-𝗠𝗼𝗧 𝗽𝗮𝗿𝗮𝗺𝗲𝘁𝗲𝗿𝘀, 𝗶𝘁 𝗱𝗲𝗹𝗶𝘃𝗲𝗿𝘀 𝗴𝗲𝗻𝗲𝗿𝗮𝘁𝗶𝗼𝗻 𝗮𝗻𝗱 𝗲𝗱𝗶𝘁𝗶𝗻𝗴 𝗾𝘂𝗮𝗹𝗶𝘁𝘆 𝘁𝗵𝗮𝘁 𝗿𝗶𝘃𝗮𝗹𝘀 𝗰𝗼𝗺𝗺𝗲𝗿𝗰𝗶𝗮𝗹 𝗰𝗹𝗼𝘀𝗲𝗱-𝘀𝗼𝘂𝗿𝗰𝗲 𝗺𝗼𝗱𝗲𝗹𝘀.

🌟This preview brings:
🔹𝗨𝗽 𝘁𝗼 4𝗞 𝗿𝗲𝘀𝗼𝗹𝘂𝘁𝗶𝗼𝗻, delivering richer details and fewer visual artifacts
🔹𝗦𝘁𝗿𝗼𝗻𝗴𝗲𝗿 𝗘𝗡/𝗖𝗡 𝘁𝗲𝘅𝘁 𝗿𝗲𝗻𝗱𝗲𝗿𝗶𝗻𝗴 𝗮𝗻𝗱 𝗰𝗼𝗺𝗽𝗹𝗲𝘅 𝗹𝗮𝘆𝗼𝘂𝘁 𝗰𝗼𝗺𝗽𝗼𝘀𝗶𝘁𝗶𝗼𝗻
🔹𝗠𝗼𝗿𝗲 𝗽𝗿𝗲𝗰𝗶𝘀𝗲 𝘃𝗶𝘀𝘂𝗮𝗹 𝗰𝗼𝗻𝘁𝗿𝗼𝗹 with long, multi-constraint prompts
🔹𝗠𝗼𝗿𝗲 𝗿𝗲𝗹𝗶𝗮𝗯𝗹𝗲 𝗻𝗮𝘁𝗶𝘃𝗲 𝗶𝗺𝗮𝗴𝗲 𝗲𝗱𝗶𝘁𝗶𝗻𝗴 across reference-guided style transfer, multi-reference composition, local infographic editing and text editing

🌟Improvements over U1 on generation & editing benchmarks:
🔹Qwen-Image-Bench: 47.14 → 55.20 (w/ PE)
🔹ImgEdit-Bench: 3.90 → 4.37
🔹GEdit-Bench-en: 7.47 → 8.17
🔹GEdit-Bench-zh: 7.42 → 8.05

From U1 to U1.5, this open-source journey reflects our continued innovation at the foundation-model level. Meanwhile, U1 Pro, our production-ready model for creators, will open for public access soon.

🛠️ GitHub: https://github.com/OpenSenseNova/SenseNova-U1/blob/feat/u1.5_preview/docs/u1.5_preview.md
🤗 Hugging Face: https://huggingface.co/sensenova/SenseNova-U1.5-8B-MoT-Preview
👾 Discord: https://discord.com/invite/BuTXPHmQub