PaliGemma 2
Google DeepMind
A lightweight open vision-language model family combining a SigLIP vision encoder with Gemma language models. It supports image and text input for captioning, visual questions, object detection and text reading, with 3B, 10B and 28B sizes. Pretrained checkpoints require task-specific tuning; mixed-task checkpoints are intended mainly for exploration and research.
What it helps with
- Fine-tune a crop image captioner
- Ask questions about field or equipment images
- Detect and localize objects in farm photos
- Read labels and documents from images
- Adapt a VLM to remote-sensing scenes
- Generate descriptions from short field videos
Scope & audience
Developer / AI builder · Scientist / researcher · Ag-tech product team
Agrifood data & AI infrastructure
Cross-commodity
Access & technical details
Free but sign up first
Model download / weights
You run it yourself
Not stated by provider
Featured media
Source information
This listing was compiled from publicly available source material. Verify current details with the provider.
Explore more like this
sgarbi/bert-fda-nutrition-ner
- Find nutrition terms inside labelling rules
- Pull ingredient wording out of regulatory text
- Tag nutrient names in food documents
- Make sense of long labelling text
- Build search over food label rules
- Structure nutrition wording from regulations
Phi-4-multimodal-instruct
- Combine field photos voice and text on one edge model
- Build a spoken crop-scouter assistant
- Inspect equipment images without a cloud call
- Translate field speech locally
- Create a multimodal farm operations interface
- Fine-tune an edge VLM for agrifood imagery
Contact us to claim this listing and add details, screenshots or a demo.