Qwen3 VL 8B Thinking

Qwen3 VL 8B Thinking is a compact 8-billion-parameter vision-language model developed by Alibaba's Qwen team, capable of processing images, documents, and video alongside text. Its 'Thinking' mode enables deliberate, step-by-step reasoning for complex visual tasks such as document analysis, OCR-heavy workflows, visual QA, and GUI-style agent interactions. Source: supabase.