Tootfinder

Alibaba releases its new Qwen3-235B-A22B-Instruct-2507 model on Hugging Face, improving on Qwen 3's reasoning, accuracy, and multilingual understanding (Carl Franzen/VentureBeat)
https://venturebeat.com/ai/alibabas-new…

Alibaba launches a free consumer-focused AI chatbot in its Quark app, powered by Qwen 3, and plans to open pre-sales for its $660 Quark AI Glasses on October 24 (Reuters)
https://www.reuters.com/world/asia-pacific/alibaba-launc…

🔧 Comparable performance to #ClaudeSONnet4 in agentic browser-use and tool-use tasks
💾 Trained on 7.5T tokens with 70% code ratio while preserving general and math abilities
⚡ Long-horizon reinforcement learning using 20,000 parallel environments on #AlibabaCloud infrastru…

TBGRecall: A Generative Retrieval Model for E-commerce Recommendation Scenarios
Zida Liang (Shanghai Jiaotong University), Changfa Wu (Alibaba Inc), Dunxian Huang (Alibaba Inc), Weiqiang Sun (Shanghai Jiaotong University), Ziyang Wang (Alibaba Inc), Yuliang Yan (Alibaba Inc), Jian Wu (Alibaba Inc), Yuning Jiang (Alibaba Inc), Bo Zheng (Alibaba Inc), Ke Chen (Alibaba Inc), Silu Zhou (Alibaba Inc), Yu Zhang (Alibaba Inc)

TBGRecall: A Generative Retrieval Model for E-commerce Recommendation Scenarios
Recommendation systems are essential tools in modern e-commerce, facilitating personalized user experiences by suggesting relevant products. Recent advancements in generative models have demonstrated potential in enhancing recommendation systems; however, these models often exhibit limitations in optimizing retrieval tasks, primarily due to their reliance on autoregressive generation mechanisms. Conventional approaches introduce sequential dependencies that impede efficient retrieval, as they a…

Filing: Shanghai-based self-driving company Banma aims to spin off from Alibaba via a Hong Kong IPO, reducing Alibaba's stake from the current 44.72% to 30% (Aileen Chuang/South China Morning Post)
https://www.scmp.com/business/ban…

Alibaba to spin off Banma self-driving business via Hong Kong IPO
Alibaba’s stake in Banma will drop to 30 per cent from the current 44.72 per cent after the IPO, the tech behemoth says.

Reduced complexities for sequences over finite alphabets
John M. Campbell, James Currie, Narad Rampersad
https://arxiv.org/abs/2509.16034 https://arxiv.org…

Reduced complexities for sequences over finite alphabets
Letting $w$ denote a finite, nonempty word, let $\text{red}(w)$ denote the word obtained from $w$ by replacing every subword $s$ of $w$ of the form $cc \cdots c$ for a given character $c$ (such that there is no character immediately to the left or right of $s$ equal to $c$) with $c$. Complexity functions for infinite words play important roles within combinatorics on words, and this leads us to introduce and investigate variants of the factor and abelian complexity functions using the given red…

Google-Mutter Alphabet ist erstmals 3 Billionen US-Dollar wert
Die Aktie des Google-Mutterkonzerns hat ein neues Rekordhoch erreicht – der US-Konzern durchbricht die Schallmauer von drei Billionen US-Dollar.
…

Alibaba releases Qwen3-Omni, a family of open-source AI models that can process text, audio, image, and video inputs and generate both text and speech outputs (Carl Franzen/Venturebeat)
https://venturebeat.com/ai/chinas-alibaba-challenges-u…

Alibaba-backed Z.ai, formerly Zhipu, launches a general-purpose AI agent app, which lets users use natural language to book hotels, order takeaway, and more (Danielle Popov/South China Morning Post)
https://www.scmp.com/tech/tech-trends/arti…

Chinese unicorn Z.ai, Alibaba Cloud team up to deploy new AI agent
The Z.ai agent is free and available on the Android and iOS mobile platforms.

Alibaba Cloud details a GPU pooling system that it claims reduced the number of Nvidia H20 required by 82% when serving dozens of LLMs of up to 72B parameters (Vincent Chow/South China Morning Post)
https://www.scmp.com/business/article/3329

Alibaba Cloud claims to slash Nvidia GPU use by 82% with new pooling system
The new Aegaeon system can serve dozens of large language models using a fraction of the GPUs previously required, potentially reshaping AI workloads.