On April 29,Watch Four Riders (Hellfighters of the East) Online Alibaba unveiled Qwen3, its newest large language model and China’s first hybrid reasoning model that integrates both fast and slow thinking modes to reduce computational costs.
The Qwen3 series includes a range of models, such as the fine-tuned Qwen3-30B-A3B and its pre-trained base, now available across major platforms. Alibaba Cloud also open-sourced two Mixture-of-Experts (MoE) models: the flagship Qwen3-235B-A22B, with over 235 billion parameters, and the lightweight Qwen3-30B-A3B, with 30 billion total and 3 billion active parameters. According to Alibaba Cloud, Qwen3-235B-A22B delivers competitive results in coding, math, and general reasoning benchmarks, rivaling top models like DeepSeek-R1, 01.AI’s o1 and o3-mini, Grok-3, and Gemini 2.5 Pro.
Qwen3’s two reasoning modes allow users to toggle between in-depth step-by-step answers or rapid responses, depending on task complexity, a flexible design aimed at balancing speed and intelligence. [Alibaba, in Chinese]
(Editor: {typename type="name"/})
Hurricane Laura's impact lingered with nightmarish mosquito swarms
Snap hires law firm to investigate racially insensitive incidents
Google Maps just made it way easier to rent a bike in 10 cities
How the internet helped two friends to confess their love for each other
Best Amazon Fire TV Cube deal: Save $30 at Amazon
Everything coming to Hulu in August 2020
Oh good, there's now a swimsuit with Donald Trump's face on it
New emoji in iOS 14 include ninja and pinched fingers
The State of 5G: When It's Coming, How Fast It Will Be & The Sci
Finally, you can buy this giant Frankenstein of a sex toy
接受PR>=1、BR>=1,流量相当,内容相关类链接。