#31 · Primary category： Foundation Models
Eagle
Eagle: Frontier Vision-Language Models with Data-Centric Strategies
Project last updated：06/24/26
GitHub Stars
3.5K
Forks
335
Contributors
12
License
Apache-2.0
Why we included this project
Researchers who want to train their own multimodal models rather than rely on a hosted API will find a complete pipeline here. The codebase combines several vision encoders with complementary strengths into a single LLM, which helps with resolution-sensitive jobs like OCR, document understanding, and chart reading. Alongside the training code, the repo includes the Eagle-SFT-1.8M instruction dataset, an evaluation harness, and a Gradio demo, so a team can go from pretraining to a tuned model without assembling pieces from elsewhere. That makes it a solid base for experimenting with fusion and data strategies in the LLaVA-style family, and a handy reference for comparing approaches.
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