#44 · Primary category： Foundation Models
Ovis
A novel Multimodal Large Language Model (MLLM) architecture, designed to structurally align visual and textual embeddings.
Project last updated：07/15/26
GitHub Stars
1.5K
Forks
89
Contributors
5
License
Apache-2.0
Why we included this project
Ovis is a practical pick for teams that need vision-language capability without gluing together separate vision and text models. It ships a full line of open multimodal LLMs from 2B up to 34B, all built around an architecture that aligns visual and textual embeddings at a structural level. The range matters because you can trade latency against capability depending on the job, and recent releases add OCR, multi-image and video input, plus a thinking mode for reasoning-heavy tasks. Since the checkpoints are openly hosted on Hugging Face with a live demo, you can test one on your own data before choosing a deployment path. It's a solid reference when you want a model that actually understands what an image contains instead of bolting vision on as an afterthought.
Articles for this project
No articles for this project yet.
To suggest a topic or contribute an article, contact us.
Related projects in this category
transformers
🤗 Transformers: the model-definition framework for state-of-the-art machine learning models in text, vision, audio, and multimodal models, for both inference and training.
CLIP
CLIP (Contrastive Language-Image Pretraining), Predict the most relevant text snippet given an image
MiniCPM-V
A Pocket-Sized MLLM for Ultra-Efficient Image and Video Understanding on Your Phone
generative-models
Generative Models by Stability AI
unilm
Large-scale Self-supervised Pre-training Across Tasks, Languages, and Modalities