Qwen

122B MoE LLM (10B active) for coding, reasoning, multimodal chat. Agent-ready.

Next-gen Qwen 3.5 VLM (400B MoE) brings advanced vision, chat, RAG, and agentic capabilities.

Qwen3-Next Instruct blends hybrid attention, sparse MoE, and stability boosts for ultra-long context AI.

Reasoning model from Alibaba with strong math and coding performance. Open-weight and cost-effective.

Versatile multilingual model with 72B parameters. Strong at code generation and multilingual tasks.

Code-specialized model matching GPT-4o on coding benchmarks. Supports 29+ programming languages.