- Maker
- Alibaba (Qwen team)
- Modality
- Text, image, and video input; text output
- Open weights
- No
- License
- Proprietary
- Context window
- 1,000K tokens
- Released
- November 16, 2026 (general availability); preceded by the qwen4-max-preview on November 11, 2026
- Parameters
- Not disclosed by Alibaba
- Architecture
- Not disclosed — Alibaba describes Qwen 4 as built on a revised underlying architecture with stronger world knowledge and multimodal perception
- Reasoning modes
- Thinking on by default; reasoning effort adjustable (low / medium / xhigh)
- Structured outputs
- JSON Object and JSON Schema, per Alibaba's model docs
- Max output
- 384K tokens
- Capabilities
- Vision, Function calling, Reasoning, Web search, Code-optimized
- Privacy on Venice
- Anonymized — prompts not stored
- Available on Venice since
- Nov 2026