A
licenseA
qualityC
maintenanceEnables text-only main models to understand images by routing local image files or URLs to a DeepSeek vision model and returning the textual analysis. Supports up to 4 images per call for description, comparison, and OCR tasks.
1
19
MIT