VITB16_IN1k is a Vision Transformer model designed for image embedding and visual grouping tasks. Developed as part of the EIDORA model zoo, it is intended for use cases such as transformer-based image embedding baselines and general image grouping and retrieval. The model is suited for deployment on recent laptops and desktops, balancing quality and domain coverage with moderate runtime requirements.
The model accepts image inputs and produces a primary output in the form of a float32 embedding tensor with shape [batch, 768]. These embeddings are already normalized, facilitating their use in downstream applications. VITB16_IN1k is not recommended for very large projects where lighter models may be preferable, nor is it intended for processing text, video, or audio inputs.
Delivery is via ONNX format, supporting use with onnxruntime, and it is available through the Hugging Face platform. The model is distributed under a BSD 3-Clause license. 11929 and is associated with the EIDORA organization.
VITB16 IN1k sits in PulseGate's Foundation models & chat category. It provides a robust image embedding model for visual grouping and retrieval tasks in AI applications. VITB16 IN1k is an open-source project aimed at computer vision researchers and developers. The project is open source (Apache-2.0). It runs on the web and the command line, and it can be self-hosted.
Behind VITB16 IN1k is EIDORA, and the product first shipped in 2020. The project is developed in the open on GitHub with 12.6k stars and 1 commits in the last 90 days. Across PulseGate's embedding index, VITB16 IN1k has few near neighbours, marking it as relatively distinct. Among its 5 catalogued features are image embedding, vision transformer, and ONNX support.
Latest indexed changes and source events
EIDORA/VITB16_IN1k discovered by the PulseGate indexer
Other apps tracked under the same category.
