Models

Open-weight experimental preview of the Qwen4 architecture: hybrid-attention MoE (125B total, 6B active) with vision encoder for coding, agent tasks, and image and video understanding

Dense 27B vision-language model for coding, agent tasks, and image and video understanding

Open-weight sparse MoE (2.4T total, 95B active), the open-weight twin of Qwen3.8 Max for coding, research, complex reasoning, and agentic workflows

Qwen vision-language model for visual reasoning, documents, and agent tasks

Open multimodal Qwen MoE for local agents that need vision, audio, and code

Qwen vision-language model for visual reasoning, documents, and agent tasks

Qwen instruction model for multilingual chat, reasoning, and tool use

Qwen vision-language model for visual reasoning, documents, and agent tasks

Qwen vision-language model for visual reasoning, documents, and agent tasks

Large open Qwen multimodal MoE for visual agents and long technical tasks

Open-weight Qwen coding model for agents, repository edits, and multi-turn tool use

Qwen vision-language thinking model for visual reasoning, documents, and agent tasks

Efficient Qwen thinking model for local reasoning, math, and coding agents

Sparse MoE Qwen model with 3B active parameters for efficient chat and reasoning

Dense open Qwen model for self-hosted chat, reasoning, and coding

Large open Qwen MoE for multilingual reasoning, coding, and tool use

Open reasoning model from the Qwen team for math, coding, and step-by-step problem solving