AI Models from Qwen

Qwen3 8B Compact Qwen3 powerhouse — everyday coding, multilingual analysis, and step-by-s... Qwen3 14B Mid-size Qwen3 — balanced speed and intelligence. Strong on coding, analysis, an... Qwen3 32B 32B dense Qwen3 — excels at complex coding, mathematics, and deep multi-step rea... Qwen3 30B-A3B (MoE) MoE architecture — 30B total with only 3B active params. Efficient & smart: near... Qwen3 48B Instruct (Jul 25) July 2025 Qwen3 48B instruct — updated instruction following, improved coding, a... Qwen3 235B-A22B (MoE) Flagship Qwen3 — 235B total, 22B active. World-class reasoning, elite coding, 10... Qwen3 235B-A22B (Jul 25) July 2025 updated Qwen3-235B — improved instruction following and enhanced capab... Qwen3 235B Thinking (Jul 25) Thinking-mode Qwen3-235B — always uses extended chain-of-thought. Best for math ... Qwen3 VL 8B Latest Qwen3 vision-language model — understands images, charts, diagrams, and s... Qwen3 VL 235B Thinking Largest Qwen vision model — 235B with extended thinking. Elite for complex visua... Qwen3 Coder Next Next-gen Qwen Coder — agentic software engineering, repository-level code unders... Qwen3 Coder Next FP8 FP8-quantized Qwen3 Coder Next — near-identical quality at lower cost and faster... Qwen3 Coder 30B-A3B MoE code specialist — 30B total, 3B active. Excellent for agentic coding workflo... Qwen3 Coder 480B Largest Qwen Coder — 480B params, 35B active, 256k context. The flagship for ful... Qwen3 Coder 480B FP8 FP8 version of the flagship Qwen3 Coder 480B — same elite capabilities at reduce... Qwen3 Next 80B-A3B Qwen3 Next 80B MoE — strong general-purpose model with 80B knowledge compressed ... Qwen3 Next 80B Thinking Thinking-mode Qwen3 Next 80B — extended chain-of-thought reasoning for math, sci... Qwen2.5 7B Instruct Previous-gen Qwen2.5 7B — reliable, well-tested, excellent for general instructi... Qwen2.5 72B Instruct Dense 72B Qwen2.5 — previous flagship, excellent across coding, writing, math, a... Qwen2.5 Coder 7B Dedicated code model trained specifically on code. Best fast option for programm...