Boogu-Image-0.1: Boosting Open-Source Unified Multimodal Understanding and Generation

2026-07-14

AI Summary

This work addresses the significant performance and efficiency gap between open-source and closed-source multimodal models in unified understanding, generation, and inference. The authors propose a unified multimodal model family—comprising Base, Turbo, Edit, and Edit-Turbo variants—that leverages high-quality image–text pair pretraining, instruction fine-tuning, a unified architecture, and an intelligent expansion strategy during inference. Trained on only 208.62 million unique images at an estimated cost of approximately $400,000, the models achieve state-of-the-art results among open-source systems across standard benchmarks, demonstrating competitive performance with leading closed-source counterparts. Capabilities include efficient text-to-image generation, rapid inference, image editing, and bilingual (Chinese–English) text rendering. The authors release all model weights, code, and training recipes to foster reproducibility and further research.

Abstract

We introduce Boogu-Image-0.1, an open-source unified multimodal understanding and generation model family, comprising Base, Turbo, Edit, and Edit-Turbo variants. It delivers competitive performance in high-quality text-to-image generation, fast inference, instruction-based editing, and bilingual (Chinese-English) text rendering. Closed-source multimodal systems like Nano-Banana-Pro and GPT-Image-2 achieve strong performance through system-level integration rather than a single model, yet their internal practices remain largely undisclosed. In this work, we demonstrate that targeted improvements in model understanding, data quality, and training pipelines, coupled with agentic inference-time scaling, can substantially enhance generation and editing performance even under highly constrained compute budgets. Comprehensive evaluations show that Boogu-Image-0.1 consistently matches or surpasses other open-source models across standard benchmarks, and achieves results approaching leading closed-source systems. Notably, this is accomplished with only 208.62 million unique images. The base model's theoretical training cost is only approximately \$400K. We share practical discussions that we believe are valuable to the broader research community, and release weights, code, and recipes under Apache 2.0 to advance the open ecosystem for unified multimodal understanding and generation. Our code is available here: https://github.com/Boogu-Project/Boogu-Image.

Problem

• multimodal understanding
• image generation
• open-source models
• instruction-based editing
• bilingual text rendering

Innovation

• unified multimodal model
• agentic inference-time scaling
• open-source image generation
• instruction-based editing
• bilingual text rendering