#30 · Primary category： Foundation Models
Skywork-R1V
Skywork-R1V is an advanced multimodal AI model series developed by Skywork AI, specializing in vision-language reasoning.
Project last updated：07/29/26
GitHub Stars
3.2K
Forks
283
Contributors
13
License
MIT
Why we included this project
The Skywork-R1V family of open-weight models targets a gap in the vision-language space: they reason step by step about an image instead of just captioning it. The series has gone through several generations, with the latest R1V3-38B trained using reinforcement learning to sharpen its chain-of-thought reasoning, and it scores well on benchmarks like MMMU and MathVista that demand real understanding. The repo keeps things practical, offering inference code for Transformers and vLLM along with AWQ-quantized checkpoints, so a single GPU with at least 30GB of memory can handle the model. That combination of strong reasoning and accessible deployment makes it a sensible pick for engineers or researchers working on document analysis, chart interpretation, or math-heavy multimodal tasks.
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