Qwen3.5-397B-A17B
MultimodalQwen3.5-397B-A17B is the flagship Qwen model with a Mixture-of-Experts architecture containing 397 billion parameters and 17 billion active. Delivers state-of-the-art performance in knowledge, reasoning, coding, math, multilingual understanding, instruction following, and agentic tasks.
Key Specifications
Timeline
Technical Specifications
Benchmark Results
Model performance metrics across various tests and benchmarks
Other Tests
License & Metadata
Compare Qwen3.5-397B-A17B
All comparisonsArticles about Qwen3.5-397B-A17B
Alibaba Doubles Down on Open Source at ModelScope DevCon
Alibaba confirmed its commitment to open-sourcing Qwen, Wan, and new model families at ModelScope DevCon in Nanjing. Here's what was announced.
Qwen 3.5 Is Alibaba's Bid to Win the Agentic AI Era
Alibaba's Qwen 3.5 family uses extreme MoE efficiency to beat models 7x its size. The flagship is now live on Arena — and claims to outperform GPT-5.2.
Similar Models
All ModelsQwen3 VL 32B Thinking
Alibaba
Qwen2-VL-72B-Instruct
Alibaba
Qwen2.5 VL 72B Instruct
Alibaba
Qwen2.5 VL 7B Instruct
Alibaba
Qwen2.5 VL 32B Instruct
Alibaba
Qwen2.5-Omni-7B
Alibaba
QvQ-72B-Preview
Alibaba
Qwen3.5 122B A10B
Alibaba
Recommendations are based on similarity of characteristics: developer organization, multimodality, parameter size, and benchmark performance. Choose a model to compare or go to the full catalog to browse all available AI models.