ByteDance-Seed/Bagel

Health v2 · maintenance only

Flags: no_releases

Adoption not part of the score

6191 stars · 544 forks observed · 2026-10-09

What it is AI-extracted, prompt v1, taxonomy v1, 2026-08-29, confidence not recorded

BAGEL is an open-source unified multimodal foundation model with 7B active parameters (14B total) trained on interleaved multimodal data. It handles both multimodal understanding and text-to-image generation, including image editing and world-modeling tasks.

Use cases

• generate images from text prompts
• understand and answer questions about images
• edit images with free-form instructions
• synthesize multiview images
• run a unified vision-language and image generation model locally

When to choose

• you need one model for both multimodal understanding and image generation
• you want an open-source alternative to specialist text-to-image models like SD3
• you need advanced image editing or world-modeling capabilities

When to avoid

• you only need lightweight image classification
• you lack GPU resources for a 14B-parameter model
• you need a production-ready hosted API rather than a research model

Facets

library · maturity active

machine-learning deep-learning llm-inference image-processing nlp artificial-intelligence large-language-models computer-vision image-processing deep-learning python multimodal vision-language-model text-to-image image-editing foundation-model world-model gpu linux

Member repositories

For agents

markdown · JSON · MCP: product_card(name="ByteDance-Seed/Bagel")