showlab/Show-o

Health v2 · maintenance only

Flags: no_releases

Adoption not part of the score

1973 stars · 93 forks observed · 2026-08-28

What it is AI-extracted, prompt v1, taxonomy v1, 2026-08-30, confidence not recorded

Show-o is a research repository implementing a unified transformer model that combines autoregressive and discrete diffusion modeling for multimodal understanding and generation. It supports vision-language tasks such as visual question answering, text-to-image generation, text-guided inpainting, and mixed-modality generation in a single model.

Use cases

• generate images from text prompts
• run visual question answering with a single model
• unify multimodal understanding and generation in one transformer
• experiment with discrete diffusion and autoregressive modeling
• do text-guided image inpainting and extrapolation
• reproduce ICLR/NeurIPS unified multimodal model research

When to choose

• you want one model handling both image understanding and generation
• you're researching unified multimodal architectures or diffusion-language hybrids
• you need a research baseline for text-to-image plus VQA in a single transformer

When to avoid

• you need a production-ready, optimized inference stack
• you only need state-of-the-art text-to-image quality from dedicated diffusion models
• you lack GPU resources for large transformer models

Facets

library · maturity active

machine-learning deep-learning image-processing nlp llm-inference artificial-intelligence machine-learning computer-vision large-language-models python multimodal diffusion-models transformer text-to-image visual-question-answering research-code unified-model natural-language-processing gpu linux

Member repositories

For agents

markdown · JSON · MCP: product_card(name="showlab/Show-o")