Facebookresearch

Rank:

Average Model Cost: $0.0076

Number of Runs: 294,217

Models by this creator

musicgen

musicgen is a model that generates music from a given prompt or melody. It uses deep learning techniques to learn the patterns and structures of music and then generates new melodies based on the input. The model can be used to compose original music, experiment with variations of existing melodies, or generate background music for videos, games, or other multimedia projects.

$-/run

245.0K

Replicate

ic_gan

facebookresearch

The Instance-Conditioned GAN (IC-GAN) is a type of image-to-image generation model that generates images conditioned on instance-level input. It takes an input instance that provides specific characteristics or attributes to guide the image generation process. IC-GAN is trained using a combination of a generator network that generates images and a discriminator network that distinguishes between real and generated images. This model can be used for various applications such as image editing and synthesis, where specific instance-level attributes are required to guide the generation process.

$0.005/run

26.3K

Replicate

detic

facebookresearch

The model named "detic" is a text-to-text model that is capable of detecting any class given class names. It has been specifically trained to identify and classify various classes based on the names provided.

$0.006/run

20.3K

Replicate

cutler

facebookresearch

The Cut and Learn model is used for unsupervised object detection and instance segmentation. It is designed to automatically discover and segment objects in images without the need for annotated training data. The model achieves this by utilizing a cut-and-paste method, where regions of the image are cut out and then reassembled in different combinations to create variations of the original image. The model then learns to predict the cut regions and uses this information to segment the objects in the image. This approach allows for the discovery and segmentation of objects in an unsupervised fashion, making it a useful tool for tasks such as image understanding and analysis.

$0.048/run

1.5K

Replicate