Draft:Every Qwen Model

```wikitext

Qwen
DeveloperAlibaba Cloud
Release2023
TypeLarge language model
LicenseMixed (Apache 2.0, proprietary, and research licenses)
Websiteqwen.ai

Qwen (Chinese: 通义千问; pinyin: Tōngyì Qiānwèn) is a family of large language models (LLMs) and multimodal artificial intelligence models developed by Alibaba Cloud. The Qwen series includes language models, coding models, reasoning models, vision-language models, audio models, and multimodal systems. Since its initial release in 2023, Qwen has become one of the largest open-weight AI model families.

History

The first generation of Qwen models was released in 2023. Subsequent generations introduced improved reasoning, multilingual support, coding ability, and multimodal capabilities.

Major releases include:

  • Qwen (2023)
  • Qwen-VL (2023)
  • Qwen1.5 (2024)
  • Qwen2 (2024)
  • Qwen2-VL (2024)
  • Qwen2-Audio (2024)
  • Qwen2.5 (2024)
  • Qwen2.5-Coder (2024)
  • Qwen2.5-Math (2024)
  • Qwen2.5-VL (2025)
  • Qwen2.5-Omni (2025)
  • QwQ-32B (2025)
  • Qwen2.5-Max (2025)
  • Qwen3 (2025)
  • Qwen3-Max (2025)

Model families

Qwen

The original Qwen family was introduced in 2023 and included models with 1.8 billion, 7 billion, 14 billion, and 72 billion parameters. These models established the foundation of the Qwen ecosystem.

Qwen-VL

Qwen-VL is a vision-language model capable of understanding both text and images. It was among Alibaba's earliest multimodal AI systems.

Qwen1.5

Qwen1.5 expanded the model lineup with additional parameter sizes ranging from 0.5 billion to 110 billion parameters. The series improved multilingual support and instruction-following capabilities.

Qwen2

Released in 2024, Qwen2 introduced major architectural improvements and improved benchmark performance. Available sizes included 0.5B, 1.5B, 7B, 57B-A14B, and 72B.

Qwen2-VL

Qwen2-VL is the second-generation vision-language family. It improved image understanding, visual reasoning, and document comprehension.

Qwen2-Audio

Qwen2-Audio added speech and audio understanding capabilities, enabling the processing of spoken language and sound-based inputs.

Qwen2.5

Qwen2.5 is a major upgrade to Qwen2. The family was trained on significantly larger datasets and improved reasoning, coding, mathematics, and long-context understanding.

Available sizes include:

  • 0.5B
  • 1.5B
  • 3B
  • 7B
  • 14B
  • 32B
  • 72B

Qwen2.5-Coder

Qwen2.5-Coder is a specialized coding model optimized for software development, code generation, debugging, and repository analysis.

Qwen2.5-Math

Qwen2.5-Math is designed for mathematical reasoning and problem solving.

Qwen2.5-VL

Released in 2025, Qwen2.5-VL improved image understanding, document parsing, chart analysis, and long-video comprehension.

Qwen2.5-Omni

Qwen2.5-Omni is a multimodal model capable of processing text, images, audio, and video inputs while generating text and audio outputs.

QwQ

QwQ is a reasoning-focused series derived from Qwen. It emphasizes chain-of-thought reasoning, mathematics, coding, and complex problem solving.

QwQ-32B

QwQ-32B is a 32-billion-parameter reasoning model released in 2025. Alibaba reported that it achieved competitive performance against much larger reasoning models.

Qwen2.5-Max

Qwen2.5-Max is a proprietary flagship model introduced in 2025. It serves as one of Alibaba's highest-performance hosted AI systems.

Qwen3

Released in April 2025, Qwen3 introduced a unified architecture supporting both reasoning ("thinking") and standard chat modes.

The open-weight lineup includes:

  • Qwen3-0.6B
  • Qwen3-1.7B
  • Qwen3-4B
  • Qwen3-8B
  • Qwen3-14B
  • Qwen3-32B
  • Qwen3-30B-A3B (Mixture-of-Experts)
  • Qwen3-235B-A22B (Mixture-of-Experts)

Qwen3 expanded support to over 100 languages and dialects.

Qwen3-Max

Qwen3-Max is Alibaba's flagship large-scale model. It reportedly contains over one trillion parameters and is optimized for advanced reasoning, coding, and AI agent tasks.

Licensing

Different Qwen models have been released under a variety of licenses, including Apache License 2.0, Qwen Research licenses, and proprietary commercial licenses.

Applications

Qwen models are used for:

  • Conversational AI
  • Software development
  • Mathematical reasoning
  • Document analysis
  • Image understanding
  • Audio processing
  • Video understanding
  • Autonomous AI agents

See also

  • Artificial intelligence
  • Large language model
  • Multimodal artificial intelligence
  • Alibaba Cloud

References

Category:Artificial intelligence Category:Large language models Category:Alibaba Group Category:Chatbots ```

Content Disclaimer

Informasi ini disarikan dari Wikipedia dan disajikan kembali untuk tujuan edukasi. Konten tersedia di bawah lisensi CC BY-SA 3.0. Kami tidak bertanggung jawab atas ketidakakuratan data yang bersumber dari kontribusi publik tersebut.

  1. The information displayed on this website is sourced in part or in whole from Wikipedia and has been adapted for the purpose of restating it. We strive to provide accurate and relevant information, however:
  2. There is no guarantee of absolute accuracy. Wikipedia is an open, collaborative project that can be edited by anyone, so information is subject to change.
  3. It is not intended to constitute professional advice. The content displayed is for informational and educational purposes only. For important decisions (e.g., medical, legal, or financial), please consult a professional.
  4. Content copyright. Wikipedia is licensed under the Creative Commons Attribution-ShareAlike License (CC BY-SA). This means that content may be reused with appropriate attribution and shared under a similar license.
  5. Responsible use. Any risk arising from the use of information from this website is entirely the responsibility of the user.