Ovis-Image Technical Report

TL;DR

Ovis-Image is a compact 7B text-to-image model optimized for high-quality text rendering, achieving performance comparable to larger models while remaining deployable on limited hardware.

Contribution

It introduces a new efficient 7B model that combines a diffusion decoder with a strong multimodal backbone and a specialized training pipeline for high-quality text rendering.

Findings

Achieves performance on par with larger models like Qwen-Image.

Operable on a single high-end GPU with moderate memory.

Demonstrates effective bilingual text rendering without oversized models.

Abstract

We introduce $\textbf{Ovis-Image}$, a 7B text-to-image model specifically optimized for high-quality text rendering, designed to operate efficiently under stringent computational constraints. Built upon our previous Ovis-U1 framework, Ovis-Image integrates a diffusion-based visual decoder with the stronger Ovis 2.5 multimodal backbone, leveraging a text-centric training pipeline that combines large-scale pre-training with carefully tailored post-training refinements. Despite its compact architecture, Ovis-Image achieves text rendering performance on par with significantly larger open models such as Qwen-Image and approaches closed-source systems like Seedream and GPT4o. Crucially, the model remains deployable on a single high-end GPU with moderate memory, narrowing the gap between frontier-level text rendering and practical deployment. Our results indicate that combining a strong multimodal backbone with a carefully designed, text-focused training recipe is sufficient to achieve reliable bilingual text rendering without resorting to oversized or proprietary models.