Conversational Work: Qwen Launches Table Agent for Direct Generation and Editing of Excel Files

Alibaba Qwen has officially launched the "Table Agent" feature. This feature allows users to directly generate, retrieve information from, and deeply edit Excel files through natural language conversations, marking a further evolution of large models from "providing text answers" to "delivering direct results."

Qwen

Comprehensive Table Processing Capabilities

The "Table Agent" breaks down the complicated process of traditional table handling. Its core capabilities cover the following three dimensions:

Zero-Barrier Information to Table Conversion: Users only need to issue commands (such as "organize the latest list of value-added tax preferential policies"), and Qwen can automatically retrieve information online and convert it into a standard Excel file for download.
Content Summarization Through Multi-Turn Conversations: The system can understand the context of multi-turn communication. For example, after discussing travel plans or study schedules, a single sentence can be used to request that the chat details be summarized into a structured table with fields such as date, budget, and notes.
Powerful Multimodal Recognition: It supports uploading PDFs, Word documents, PPTs, even hand-drawn class schedules or photos of paper reports. Qwen can accurately recognize the content of images and text and restore them into Excel format. It also supports instruction-based operations on existing tables, such as "center alignment" or "ranking by sales revenue."

Technical Principles: Independent Sandboxes and Intelligent Planning

In terms of technical implementation, Qwen breaks down complex user requirements into an Agent execution chain. The system runs code (Coding) in an independent sandbox environment, thereby generating professional files with real formulas, complex formatting, and strict data logic. When the existing information is insufficient to support table generation, the Agent will automatically trigger online retrieval to supplement the data.

Currently, the Qwen Table Agent is available across the Qwen App, web version, and PC client, and all users can experience this productivity tool for free.

Through this feature iteration, Alibaba Qwen aims to further lower the office threshold, making Excel processing no longer dependent on complex functions, but instead becoming a simple task that can be completed by just speaking.

GAC Unveils the New Star River Smart Cabin Architecture: Deep Integration with Qwen, Creating a Mobile Third Space Where One Command Completes Everything

GAC launches the Star River Smart Cabin ADiGO Intelligence architecture, deeply integrating Alibaba's Qwen large model to revolutionize the smart cabin experience. Based on real data from over 3.75 million vehicles, this architecture achieves rapid response within 1.6 seconds, an accuracy rate of over 95%, and proactive scenario-based services through three core elements: multimodal perception, emotional interaction, and scenario engines, driving human-vehicle interaction towards greater intelligence and emotionalization.

215 SOTA Wins! Alibaba Releases Qwen3.5-Omni: Outstanding Cross-Modal Performance Exceeding Gemini

Alibaba released the new generation cross-modal large model Qwen3.5-Omni, achieving seamless cross-modal interaction with images, videos, voice, and text. It performed excellently in 215 tasks, marking a significant breakthrough for domestic large models in the field of audiovisual interaction, reaching global top levels.

Qwen Enters the Hongqi Car AI Assistant, Starting to Work in the Cabin

The Alibaba AI assistant Qwen has been integrated into the intelligent cabin of Hongqi cars, and will be first launched in the Hongqi HS6 PHEV. Users can perform multi-task processing through voice commands, such as planning a navigation route, recommending restaurants along the way, and arranging schedule time. This marks Qwen's expansion from AI glasses to more complex in-car scenarios, accelerating its entry into the physical world.

Just one sentence to call a taxi! Alibaba Qwen launches AI assistance: Local life traffic is about to be reshuffled

Alibaba's Qwen AI assistant has launched the 'AI Taxi' feature, allowing users to request a ride and plan their journey through natural language conversation. This move is not only a technological upgrade but also a key strategy for Alibaba to reshape the local life traffic entrance with AI-native services, showcasing its accelerated strategy toward becoming a comprehensive assistant.

Core Members of Qwen Move to ByteDance: The Competition for Large Model Talent Heats Up Again

Alibaba's Tongyi Lab has recently undergone restructuring, with the original Qwen team being split, leading to talent turnover. After Lin Jinyang left, Yu Bowen, the former head of the large model pre-training team of Qwen, also joined ByteDance, taking on the role of head of the pre-training team for the Seed team's visual model and multimodal interaction team. This reflects the intensifying competition for talent in the field of large models in China, and the industry structure is undergoing a new round of reshaping.