Ovis-Image-7B

The model, with 7 billion parameters, was developed for generating high-quality images with a focus on precise text representation, while optimized for operation on limited computational resources. Built on its predecessor, Ovis-U1, it is designed to operate efficiently on a single high-performance GPU.

Key features:

• High accuracy in text rendering under dense element layouts (e.g., posters, logos, UI designs, infographics). The generation preserves readability, orthographic correctness, and semantic coherence of the text, accounting for diverse fonts, sizes, and aspect ratios.
• Text quality comparable to 20B-sized models (e.g., Qwen-Image), while utilizing a smaller memory footprint.

---

The model is a component of the image generation pipeline, consisting of:

• Text encoder: ~2B parameters,
• Transformer: ~7B parameters,
• VAE: ~83 parameters,

Total: ~9B parameters

Announce Date: 28.11.2025
Parameters: 7B
Developer: AIDC-AI
Diffusers Version: 0.36.0.dev0
vLLM-Omni Version: 0.12.0rc1
License: Apache 2.0

Use our pre-built public endpoints for free to test inference and explore Ovis-Image-7B capabilities. You can obtain an API access token on the token management page after registration and verification.

Rent your own physically dedicated instance with hourly or long-term monthly billing.

We recommend deploying private instances in the following scenarios:

• maximize endpoint performance,
• enable full context for long sequences,
• ensure top-tier security for data processing in an isolated, dedicated environment,
• use custom weights, such as fine-tuned models or LoRA adapters.

Contact our dedicated neural networks support team at nn@immers.cloud or send your request to the sales department at sale@immers.cloud.