China AI Native Industry Insights – 20251013 – Alibaba | Gaga AI | Tencent | more

Explore Alibaba’s enhanced Qwen Code with Plan Mode, Vision Boost, and Zed integration for safer, faster coding. Dive into GAGA-1’s groundbreaking impact on AI video acting. Experience Tencent’s open-source SongBloom model revolutionizing AI music generation. Discover more in Today’s China AI Native Industry Insights.

1. Alibaba’s Qwen Code just got smarter: Plan Mode, Vision Boost, Zed integration. Build safer, code faster.

🔑 Key Details:
– Plan Mode: Qwen Code introduces Plan Mode, where the AI first generates a full implementation plan for approval before making any code changes.
– Vision Intelligence: Automatically switches to Qwen3-VL-Plus when image input is detected—even in YOLO mode—with support for 256K input and 32K output tokens.
– Zed Integration: Now supports both OpenAI and Qwen OAuth authentication within the Zed code editor.
– Fine-Grained Control: Users can toggle loop detection, confirm file overwrites (e.g., QWEN.md), and adjust agent workflow behavior.

💡 How It Helps:
– Developers: Gain more visibility and control before code execution, reducing risk of unintended changes.
– AI Engineers: Leverage multimodal capabilities to handle visual data alongside code.
– Product Teams: Integrate with Zed for smoother workflows using secure OAuth-based authentication.

🌟 Why It Matters:
With these updates, Alibaba’s Qwen Code evolves into a more controllable and autonomous coding agent. Plan Mode ensures transparency and safety, while enhanced visual support and editor integration make it practical for real-world software development. These features bring Qwen Code closer to a production-ready AI pair programmer.

Original Chinese article: https://x.com/Alibaba_Qwen/status/1976933486654374240

English translation via free online service: https://translate.google.com/translate?hl=en&sl=zh-CN&tl=en&u=https%3A%2F%2Fx.com%2FAlibaba_Qwen%2Fstatus%2F1976933486654374240

Video Credit: Qwen

2. GAGA-1: Revolutionizing the Future of AI Video Acting

🔑 Key Details:
– Breakthrough Model: GAGA-1 is a holistic AI actor, merging voice, lipsync, and expressions into a single coherent performance.
– Overcoming Limitations: It addresses common flaws in traditional AI video generators, such as disjointed lip sync and lack of emotional depth.
– Versatile Applications: GAGA-1 can be utilized across industries, including content creation, education, and corporate communication.

💡 How It Helps:
– Content Creators: Allows for authentic video content production more efficiently, enhancing audience engagement.
– Marketers: Facilitates the creation of engaging ad campaigns with customizable digital actors.
– Educators: Engages learners through lifelike virtual instructors, making education more immersive.

🌟 Why It Matters:
GAGA-1’s introduction marks a transformative shift in digital performance, blurring the lines between human and AI actors. By streamlining video production, it positions itself competitively in the content creation landscape, promising richer, more engaging experiences while addressing the persistent challenges of previous models. This advancement could redefine audience interaction in various sectors.

Original Chinese article: https://gaga.art/blog/gaga-1/

English translation via free online service: https://translate.google.com/translate?hl=en&sl=zh-CN&tl=en&u=https%3A%2F%2Fgaga.art%2Fblog%2Fgaga-1%2F

Video Credit: Gaga AI

3. Tencent Releases Open-Source SongBloom Model for AI Music Generation

🔑 Key Details:
– Tencent AI Lab, in collaboration with universities, has launched SongBloom, an open-source song generation model.
– The model can produce high-quality, 2.5-minute songs from just 10 seconds of a sample and lyrics.
– SongBloom outperforms existing models in audio quality and lyric accuracy while achieving near state-of-the-art performance in music generation.

💡 How It Helps:
– Music Creators: Access to a powerful tool that improves song quality and aligns lyrics with melodies, streamlining the creative process.
– AI Developers: The open-source nature of SongBloom offers opportunities for innovation and customization in music AI research.

🌟 Why It Matters:
SongBloom’s innovative use of self-regressive diffusion models signals a major advancement in AI music technology, addressing previous limitations in song generation accuracy and quality. This leap fosters more realistic and creative outputs, positioning Tencent at the forefront of the AI music landscape, enhancing competition and encouraging new industry standards.

Original Chinese article: https://mp.weixin.qq.com/s/xiiLU9cz33PTpPNSh728Ig

English translation via free online service: https://translate.google.com/translate?hl=en&sl=zh-CN&tl=en&u=https%3A%2F%2Fmp.weixin.qq.com%2Fs%2FxiiLU9cz33PTpPNSh728Ig

Video Credit: The original article

That’s all for today’s China AI Native Industry Insights. Join us at AI Native Foundation Membership Dashboard for the latest insights on AI Native, or follow our linkedin account at AI Native Foundation and our twitter account at AINativeF.