Qwen (Tongyi Qianwen)
Qwen (Tongyi Qianwen) is a series of large language models developed by Alibaba Group, with capabilities encompassing natural language understanding and generation, multi-turn dialogue, text summarization, and code writing. The series includes model versions of varying parameter scales; some models are released as open source, supporting local deployment and secondary development by developers. The models are widely applied in intelligent assistants, content generation, and enterprise intelligence scenarios.
Core Facts
Qwen-Image-Edit is an image editing model launched by Alibaba's Tongyi Qianwen team, with capabilities for image generation and local editing.
Qwen Image is a text-to-image model launched by Alibaba.
WorkBuddy's interface is very similar to those of Qianwen, Doubao, and GPT.
Timeline (last 90 days)
Conversational large models such as Doubao, DeepSeek, and Tongyi Qianwen can understand needs, analyze problems, and give solutions, but cannot directly operate the user's computer to execute tasks.
Alibaba released the Tongyi Qianwen series of open-source large models in August 2023, launching versions of different scales such as 7B and 72B, and has now upgraded to Tongyi Qianwen 2, which possesses multimodal interaction capabilities including natural language processing, visual understanding, audio processing, and AI Agent.
The Tongyi Qianwen Flash edition starts at $0.03 and $0.13, and GLM scored 1,647 points on GDPVal.
The domestic platforms listed in Pi's official reference table include DeepSeek, Kimi, MiniMax, Alibaba's Tongyi Qianwen, Xiaomi's models, and others.
General-purpose large models are trained on historical data and inherently lack internet connectivity, the ability to call external tools, conversational memory, and access to enterprise-internal proprietary data.
Ninfer is an inference engine designed specifically for Qwen-class models and has approximately a twofold speed advantage over general-purpose solutions such as llama.cpp.
Qwen Image 2.1 Turbo requires only 8 denoising steps to generate images, reducing sampling steps by 80% compared to the original version's approximately 40 steps.
After the advent of GPT, large models such as ERNIE Bot, Tongyi Qianwen, Xunfei Xinghuo, and Doubao emerged domestically in China.
OpenHands added LLM provider selection capability in V1.16, allowing users to switch between model sources such as Tongyi Qianwen, Kimi, Zhipu GLM, and DeepSeek in the settings.
The application integrates three major domestic models — ERNIE Bot, Tongyi Qianwen, and DeepSeek — managed through a unified AI Service Layer.
40 more timeline events
All Facts (20)
Qwen-Image-Edit is an image editing model launched by Alibaba's Tongyi Qianwen team, with capabilities for image generation and local editing.
90%VerifiedQwen Image is a text-to-image model launched by Alibaba.
90%VerifiedWorkBuddy's interface is very similar to those of Qianwen, Doubao, and GPT.
80%VerifiedQwen 3.8 Flash Next adopts a Mixture of Experts (MoE) architecture.
80%VerifiedTongyi Lingma is an AI programming tool developed by Alibaba based on the Tongyi Qianwen large model, deeply integrated into VS Code and JetBrains series IDEs.
80%VerifiedSpring AI decouples upper-level business code from specific model implementations by defining unified core abstract interfaces such as ChatModel and EmbeddingModel.
80%VerifiedQwen（通义千问）是阿里巴巴达摩院开源的系列大模型，覆盖从0.5B到72B的多种参数规模，支持多语言和多模态任务
80%VerifiedThe Qwen series is a family of open-source large language models developed by Alibaba Cloud's Tongyi Lab
80%VerifiedSpring AI Alibaba是阿里巴巴基于Spring AI扩展的框架，支持智能体开发
80%VerifiedMainstream domestic large models such as DeepSeek and Tongyi Qianwen all provide interface standards compatible with the OpenAI API format.
80%VerifiedWorkBuddy's interface is very similar to those of Tongyi Qianwen, Doubao, GPT, and similar tools.
75%VerifiedMainstream domestic large models include Alibaba's Tongyi Qianwen (open-source), Baidu's Wenxin series, and DeepSeek.
75%VerifiedTongyi Qianwen is a large model under Alibaba Cloud that offers an open-source version.
75%VerifiedThe approximately 20:1 ratio between Qwen3 Next Flash's 125 billion total parameters and 6 billion activated parameters represents a fairly aggressive sparsity design.
70%VerifiedQwen (the Tongyi Qwen series) is a family of open-source large models launched by Alibaba.
65%VerifiedTraditional AI dialogue tools (such as ChatGPT, Doubao, and Tongyi Qianwen) cannot directly operate local files and typically return results in text form.
65%VerifiedThe core of using WorkBuddy is to shift from the mindset of 'asking AI questions' to the mindset of 'managing AI to do the work'.
65%VerifiedDoubao and Tongyi Qianwen are widely used AI chat tools.
65%VerifiedQwen is the multimodal version of Alibaba's Tongyi Qianwen series, performing robustly on bilingual Chinese-English comprehension and visual tasks.
65%VerifiedCodex is essentially a connectivity-layer tool that can interface with multiple models, including DeepSeek, GPT, Gemini, Qwen, and other domestic and international models.
65%