We introduce Boogu-Image-0.1, an open-source unified multimodal understanding and generation model family, comprising Base, Turbo, Edit, and Edit-Turbo variants. This model delivers competitive performance in high-quality text-to-image generation, fast inference, instruction-based editing, and bilingual (Chinese-English) text rendering.
In contrast to closed-source multimodal systems like Nano-Banana-Pro and GPT-Image-2, which achieve strong performance through system-level integration rather than a single model, their internal practices remain largely undisclosed. Our work demonstrates that targeted improvements in model understanding, data quality, and training pipelines, along with agentic inference-time scaling, can significantly 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, achieving results close to leading closed-source systems. Notably, this is accomplished with only 208.62 million unique images, and the base model's theoretical training cost is 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: Boogu-Image GitHub.
Blogger's Review: The launch of Boogu-Image-0.1 marks a significant advancement in the open-source multimodal model domain, offering an efficient balance between performance and cost, which opens up new possibilities for researchers, especially in resource-constrained scenarios.