Browse deepinfra models:

All categories and models you can try out and directly use in deepinfra:


Zero-shot image classification is a powerful technique in machine learning that allows you to classify images into categories that a model has never seen before during training. This is especially useful for image classification tasks where obtaining labeled training data for every possible category is difficult or expensive. This is often the case in a variety of industries, such as healthcare, manufacturing, and e-commerce.

To build a zero-shot image classification model, you can use a technique called transfer learning, where a pre-trained model is fine-tuned on a smaller dataset with specific categories. The pre-trained model is typically trained on a large dataset of images with generic labels, such as ImageNet, which contains over a million images labeled with 1000 categories.

During the fine-tuning process, the model learns to recognize visual features that are common across different categories, such as shapes, textures, and colors. To make zero-shot predictions, the model uses a set of attributes or features that are associated with each category.

However, it's important to note that zero-shot models can sometimes struggle with fine-grained distinctions between similar categories, and may require additional training data to improve their accuracy. In these cases, you may want to consider using semi-supervised or unsupervised learning techniques to augment your zero-shot model with additional labeled or unlabeled data.

openai/clip-vit-base-patch32 cover image
$0.0005 / sec
  • zero-shot-image-classification

The CLIP model was developed by OpenAI to investigate the robustness of computer vision models. It uses a Vision Transformer architecture and was trained on a large dataset of image-caption pairs. The model shows promise in various computer vision tasks but also has limitations, including difficulties with fine-grained classification and potential biases in certain applications.

openai/clip-vit-large-patch14 cover image
$0.0005 / sec
  • zero-shot-image-classification

Zero-shot image classification model

openai/clip-vit-large-patch14-336 cover image
$0.0005 / sec
  • zero-shot-image-classification

A zero-shot-image-classification model released by OpenAI. The clip-vit-large-patch14-336 model was trained from scratch on an unknown dataset and achieves unspecified results on the evaluation set. The model's intended uses and limitations, as well as its training and evaluation data, are not provided. The training procedure used an unknown optimizer and precision, and the framework versions included Transformers 4.21.3, TensorFlow 2.8.2, and Tokenizers 0.12.1.