This week, Double 11 garnered stellar performance across Taobao and Tmall, where 284 brands surpassed RMB100 million in sales as of October 24. Categories like beauty, fashion, consumer electronics, and home goods saw exceptional growth, with numerous brands doubling or even quintupling their sales compared to last year. Also, by observing user purchasing behaviors and merchant sales performance during the first round of Double 11, we have identified several exciting consumer trends, like toys and plushies’ sales surge with the rise of emotional Shopping.
Alicast returns with insights from Kaifu Zhang, Alibaba’s VP of International Digital Commerce, on how AI tools such as chatbots and smart translation are empowering small businesses to thrive globally.
Elsewhere on the AI front, Alibaba Cloud’s Qwen 2.5 topped the OpenCompass leaderboard, becoming the first open-source LLM to outperform top-tier proprietary models.
In other news, DAMO Academy’s AI-driven research unlocks new opportunities in crop breeding by leveraging DNA methylation, highlighting the academy’s focus on smart agriculture and fundamental research.
Significant Double 11 Performance in First Check-Out Across Categories on Taobao and Tmall
During the first round of Double 11 sales, Taobao and Tmall witnessed solid results across various sectors. In the initial four hours of the first checkout period, 174 brands exceeded RMB100 million in sales, with over 12,000 brands doubling their sales compared to the same period last year.
Beauty products saw exceptional performance, with 40 brands generating over RMB100 million in sales and 17 individual products surpassing this mark. Notably, new products accounted for 395 of the top 1,000 best-selling beauty items, marking a 72% increase from last year.
Apparel brands also experienced robust growth, with 3,829 brands doubling their sales compared to the previous year, including 2,083 that tripled and 1,561 that quintupled their sales. The sports and outdoor category saw 15 brands exceed RMB100 million in sales, while 52 others surpassed RMB10 million.
In consumer electronics, 12 brands—including Apple, Huawei, and Xiaomi—achieved RMB100 million in sales, and 42 brands doubled their performance. The home appliances and furniture sector saw 28 brands break the RMB100 million mark, with 39 emerging sub-categories surpassing the same threshold.
Additionally, 15 pet brands reached RMB10 million in sales within the first hour, and over 650 brands doubled their sales and orders within four hours, maintaining the segment’s high growth trajectory and strong potential in the future.
Stay tuned continuously updating on Liveblog of Double 11 2024
Alicast is back, and this time it’s all about AI!
Join Kaifu Zhang, Vice President of Alibaba International Digital Commerce, as he reveals how cutting-edge AI tools—like smart translation services and chatbots—are transforming global commerce. Discover how these innovations empower small businesses to thrive across borders, breaking down barriers and reshaping the future of e-commerce. Don’t miss this insightful conversation about the evolving role of AI in global business.
🎧 Listen now on Apple Podcast for Mainland China listeners, for Others, or on Spotify
Alibaba Cloud’s Qwen 2.5 Tops OpenCompass LLM Leaderboard as the First Open-Source Champion
According to its latest September update, Alibaba Cloud’s open-source Qwen 2.5-72B-Instruct has claimed the top spot on the OpenCompass large language model leaderboard. In various benchmarks,it surpasses even closed-source SOTA models, such as Claude 3.5 and GPT-4o.
Qwen 2.5-72B-Instruct showcased strong overall capabilities, achieving the highest score of 74.2 in coding and an impressive 77 in mathematics, outperforming Claude 3.5 (72.1) and GPT-4o (70.6). In a recent article, OpenCompass commended Qwen 2.5 as its first-ever open-source champion, reflecting the rapid progress in the open-source LLM community.
The bi-monthly leaderboard, established by the Shanghai Artificial Intelligence Laboratory, assesses LLM performance across language, knowledge, reasoning, mathematics, coding, instruction, and agent capabilities.
During the Apsara Conference 2024, Alibaba Cloud introduced over 100 open-source Qwen2.5 multimodal models to the global open-source community. These models range from 0.5 to 72 billion parameters in size and support over 29 languages.
See more about 100 open-sourced Qwen2.5 multimodal models on Apsara 2024
AI-Driven Breakthrough in Crop Breeding: DAMO Academy and Partners Unlock the Power of DNA Methylation
Recent research conducted by Alibaba DAMO Academy, Zhejiang University, and the Chinese Academy of Agricultural Sciences (CAAS) demonstrates that DNA methylation data can serve as a valuable resource for crop breeding, offering new avenues to accelerate crop improvement. The findings were published in Cell Research, a journal under the Nature portfolio.
To better understand the significance of DNA methylation, the research team leveraged advanced AI technologies, including a deep learning model named Deep Functional DNA Methylation Loci (DeepFDML). By analyzing high-quality methylomes, transcriptomes, and genomes from crop fibers, the study identified more than 287 million single methylation polymorphisms (SMPs)—the largest dataset of its kind to date.
Additionally, the study identified 43 potential genes involved in fiber development, providing a valuable list of candidate genes to guide future breeding efforts.
This breakthrough highlights the transformative potential of AI in life sciences and reinforces DAMO Academy’s long-term commitment to advancing fundamental research, including a particular focus on smart agriculture.
