On April 29,Watch Angel Has Fallen 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"/})
Sony launches new flagship XM6 headphones: Order them now
Model shares photo of her cellulite on Instagram to show 'how bloody natural it is'
Extremely happy elephant is super excited to meet the Queen
Kendrick Lamar reveals the tracklist for his new album, which features Rihanna and U2. Yes, U2.
Instagram Direct update is Facebook's latest attack on Snapchat
'Fawlty Towers' fans, rejoice: John Cleese is returning to the BBC for a new sitcom
Want your matches to message you back? These are the emojis you need.
NYT mini crossword answers for May 12, 2025
YouTube's biggest star is trying out Twitch
The cicadas aren't invading the U.S.
Instagram Direct update is Facebook's latest attack on Snapchat
接受PR>=1、BR>=1,流量相当,内容相关类链接。