Vision Medium
Balanced multimodal model pairing a 1M-token context window with deeper reasoning for analysis, content creation, and tool use across text, image, audio, video, and PDF inputs.
vispark/vision-medium- Lab
- Vispark
- Family
- -
- Providers
- 1
- Context
- 1,000,000
- Output limit
- 65,536
- Knowledge
- -
- Release
- 2024-05-15
- Updated
- 2026-09
- Weights
- Closed
- Input
- Output types
- Capabilities
- tools, reasoning, structured, temperature