AI BriefingQwen news
The latest AI and developer news from Qwen, with short summaries.
Visit website →Context length expanded to 1 million tokens
Qwen2.5-Turbo has expanded its context length to 1 million tokens and significantly improved inference speed and cost efficiency.
Qwen2.5-Coder Series: Strong Performance, Diverse Models, High Practicality
The Qwen2.5-Coder series, featuring GPT-4o-level coding performance, has been released as open source in a variety of model sizes.
Qwen2.5: A Feast of Foundation Models
Alibaba unveiled the Qwen2.5 model series, trained on 18 trillion tokens, delivering powerful performance.
Qwen2.5-LLM: Expanding the Boundaries of LLMs
Alibaba unveiled the Qwen2.5 series, trained on 18 trillion tokens to substantially boost performance.
Qwen2.5-Coder: More Coding, More Training!
The next-generation open source model **Qwen2.5-Coder**, trained on 5.5 trillion tokens to dramatically strengthen coding and math capabilities, has been released.
Qwen2.5-Math: A World-Class Open Source Math-Specialized LLM
Qwen has open-sourced the Qwen2.5-Math series, which supports CoT and TIR to enhance mathematical problem-solving ability.
Qwen2-VL: Understanding the World More Clearly
Qwen2-VL, the latest vision-language model in the Qwen model series, has been released with enhanced video understanding and Agent capabilities.
Qwen2-Audio: Talking with Your Voice
Qwen's next-generation audio model, **Qwen2-Audio**, has been released, supporting both voice and text input.
Introducing Qwen2-Math
Qwen has released Qwen2-Math, a series of math-specialized LLMs that maximize mathematical problem-solving ability.
Introducing Qwen2
Qwen has released Qwen2, its next-generation model with significantly enhanced performance and multilingual capabilities.
Extending LLM Context from 8k to 1M Using Qwen-Agent
It presents a strategy for building an agent that handles 1M context using an 8k context model, and using this to train long-context models.
Notes on Qwen-Max-0428
The Qwen team has released a large-scale model, **Qwen-Max-0428**, which surpasses the performance of the existing Qwen1.5-110B-Chat.
Qwen1.5-110B: The First 100B+ Model in the Qwen1.5 Series
The Qwen team has released Qwen1.5-110B, a 100B-scale model with performance on par with Llama-3-70B.
Coding with CodeQwen1.5
Qwen has released CodeQwen1.5-7B, an open-source code-specialized LLM trained on 3 trillion code tokens.
Qwen1.5-32B: The Pinnacle of the Qwen1.5 Language Model Series
Qwen has released the 32B-scale Qwen1.5 model, which offers the optimal balance of performance and efficiency.
Qwen1.5-MoE: Achieving 7B Model Performance with Only 1/3 of the Activated Parameters
Qwen1.5-MoE-A2.7B achieves performance on par with 7B models while reducing activated parameters to 1/3.
Introducing Qwen1.5
Qwen1.5, the next-generation model in the Qwen series, has been released with various sizes and improved performance.
Introducing Qwen-VL
Qwen has released Qwen-VL-Plus and Qwen-VL-Max, multimodal models with greatly enhanced image reasoning and recognition capabilities.
Introducing the Qwen Series
The **Qwen** series, aiming at AGI, has been released, including open-source **LLMs** and **LMMs** of various sizes.
OFASys: Implementing Multitask Learning with Just One Line of Code
OFASys is an AI framework that implements complex multimodal multitask learning with just one line of code through the 'Instruction' interface.