SigLIP 2
Google DeepMind
A multilingual vision-language encoder family for zero-shot image classification, image-text retrieval, localization and dense visual features. The downloadable checkpoints can also serve as the vision encoder inside a VLM. It is general-purpose and not agrifood-tuned.
QUICK INFO
SOLUTION TYPE
Models, data & tools
TECHNOLOGY CATEGORY
Models & fine-tunes
AREA OF WORK
Agrifood data & AI infrastructure
ACCESS
Free · Open source
DEPLOYMENT
Self-hosted / on-premise · Edge / local · Downloadable
SOURCE REVIEWED
August 26, 2026
What it helps with
- Prototype or evaluate zero-shot crop or weed image classification
- Prototype text search over farm photos
- Generate image-text embeddings for field imagery
- Evaluate a vision encoder inside an agrifood VLM
- Explore object localization in agricultural images
- Fine-tune dense visual features for crop tasks
Scope & audience
WHO IT'S FOR
Developer / AI builder · Scientist / researcher · Ag-tech product team
AREA OF WORK
Agrifood data & AI infrastructure
COMMODITY / SCOPE
Cross-commodity
Access & technical details
ACCESS
Anyone can get it
INTERFACE
Model download / weights
DEPLOYMENT
You run it yourself
LANGUAGES
Multilingual
Featured media
Source information
This listing was compiled from publicly available source material. Verify current details with the provider.
PROVIDER DOCUMENTATION
Reviewed August 2026
Explore more like this
plant-genomic-jamba
Publisher not stated
Models, data & tools
Different model designs suit different genome questions, and the options for plants are limited. This is a downloadable plant genome model using a hybrid design, trained on sequence from many plant species.
Free
Self-hosted / on-premise · Downloadable
Models & fine-tunes
plant-genomic-jamba
research-develop
research-and-scientific-knowledge
buildable-or-downloadable-component
models-and-fine-tunes
Free
Self-hosted / on-premise · Downloadable
August 23, 2026
- Plant genome model with a hybrid design
- Download a plant DNA model
- Try a different design for genome questions
- Model trained on many plant species
- Plant sequence model to fine tune
Cropformer
Chinese Academy of Agricultural Sciences / co-authors (Wang, Yan, Fang)
Models, data & tools
Predicting how a plant will perform from its genetic markers usually gives you a number with no explanation. This is code and a web tool for crop trait prediction that also points to which markers drove the result.
Free
Self-hosted / on-premise · Downloadable
Models & fine-tunes
Cropformer
research-develop
seed-genetics-and-breeding
buildable-or-downloadable-component
models-and-fine-tunes
Free
Self-hosted / on-premise · Downloadable
August 23, 2026
- Predict crop traits from genetic markers
- Which markers drove the prediction
- Explain a trait prediction result
- Run crop trait prediction in a browser
- Code for crop trait prediction
Is this your Resource?
Contact us to claim this listing and add details, screenshots or a demo.