metadata

datasets:
  - ffurfaro/PixelBytes-PokemonAll
language: en
library_name: pytorch
license: mit
pipeline_tag: text-to-image
tags:
  - image-generation
  - text-generation
  - autio-generation
  - multimodal

PixelBytes: Unified Multimodal Generation

Welcome to the PixelBytes repository! This project features models designed to generate text, audio and images simultaneously, pixel by pixel, using a unified embedding. (only testing weight)

Overview

Key Concepts

Image Transformer: Generates images pixel by pixel.
Bi-Mamba+: A bidirectional model for time series prediction.
MambaByte: A selective state-space model without tokens.

The PixelByte model generates mixed sequences of text and images, handling transitions with line breaks and maintaining image dimension consistency.

Dataset

We use the PixelBytes-PokemonAll dataset, available on Hugging Face: PixelBytes-PokemonAll. It contains text and image sequences of Pokémon for training our model.

Models Trained

3 LSTM Models: 2 Auto-regressive and 1 only predictive.

Citation

Furfaro, F. (2024). PixelBytes: A Unified Multimodal Representation Learning Project. (https://github.com/fabienfrfr/PixelBytes)

Thank you for exploring PixelBytes! We hope this model aids your multimodal generation projects.