Top AI Repos — open-source AI, indexed and scored
Top AI Repos tracks AI repositories on GitHub and answers two different questions about each one: is it moving right now, and would you bet a product on it.
Top AI Repos tracks AI repositories on GitHub and answers two different questions about each one: is it moving right now, and would you bet a product on it.
Parsing-free RAG supported by VLMs
| Date | Stars |
|---|---|
| 2026-07-24 | 972 |
| 2026-07-25 | 973 |
| 2026-07-28 | 975 |
| 2026-07-30 | 975 |
| 2026-08-06 | 975 |
Today
— stars today
This week
— stars this week
This month
— stars this month
Momentum
0.0
growth rate 0.00%/day
# VisRAG 2.0: Evidence-Guided Multi-Image Reasoning in Visual Retrieval-Augmented Generation [](https://github.com/OpenBMB/VisRAG) [](https://arxiv.org/abs/2510.09733) [](https://arxiv.org/abs/2410.10594) [](https://huggingface.co/openbmb/EVisRAG-7B) [](https://huggingface.co/openbmb/EVisRAG-3B) [](https://huggingface.co/openbmb/VisRAG-Ret) [](https://huggingface.co/collections/openbmb/visrag-6717bbfb471bb018a49f1c69) [](https://huggingface.co/spaces/tcy6/VisRAG_Pipeline) [](https://colab.research.google.com/drive/11KV9adDNXPfHiuFAfXNOvtYJKcyR8JZH?usp=sharing) <p align="center">• <a href="#-introduction"> 📖 Introduction </a> • <a href="#-news">🎉 News</a> • <a href="#-visrag-pipeline">✨ VisRAG Pipeline</a> • <a href="#%EF%B8%8F-setup">⚙️ Setup</a> • <a href="#%EF%B8%8F-training">⚡️ Training</a> </p> <p align="center">• <a href="#-evaluation">📃 Evaluation</a> • <a href="#-usage">🔧 Usage</a> • <a href="#-license">📄 Lisense</a> • <a href="#-contact">📧 Contact</a> • <a href="#-star-history">📈 Star History</a> </p> # 📖 Introduction **EVisRAG (VisRAG 2.0)** is an evidence-guided Vision Retrieval-augmented Generation framework that equips VLMs for multi-image questions by first linguistically observing retrieved images to collect per-image evidence, then reasoning over those cues to answer. **EVisRAG** trains with Reward-Scoped GRPO, applying fine-grained token-level rewards to jointly optimize visual perception and reasoning. <p align="center"><img width=800 src="assets/evisrag.png"/></p> **VisRAG** is a novel vision-language model (VLM)-based RAG pipeline. In this pipeline, instead of first parsing the document to obtain text, the document is directly embedded using a VLM as an image and then retrieved to enhance the generation of a VLM. Compared to traditional text-based RAG, **VisRAG** maximizes the retention and utilization of the data information in the original documents, eliminating the information loss introduced during the parsing process. <p align="center"><img width=800 src="assets/main_figure.png"/></p> # 🎉 News * 20251207: Released all [benchmarks](https://huggingface.co/collections/openbmb/visrag) on HuggingFace. * 20251118: Both EVisRAG and VisRAG can be easily reproduced within [UltraRAG v2](https://github.com/OpenBMB/UltraRAG). * 20251022: We upload all evaluation benchmarks in [VisRAG Collections](https://huggingface.co/collections/openbmb/visrag-6717bbfb471bb018a49f1c69) * 20251014: Released [EVisRAG-3B](https://huggingface.co/openbmb/EVisRAG-3B) on HuggingFace. * 20251014: Released **EVisRAG (VisRAG 2.0)**, an end-to-end Vision-Language Model. Released our [Paper](https://arxiv.org/abs/2510.09733) on arXiv. Released our [Model](https://huggingface.co/openbmb/EVisRAG-7B) on HuggingFace. Released our [Code](https://github.com/OpenBMB/VisRAG) on GitHub * 20241111: Released our [VisRAG Pipeline](https://github.com/OpenBMB/VisRAG/tree/master/visrag_scripts/demo/visrag_pipeline) on GitHub, now supporting visual understanding acr
Excerpt of 18,866 characters
Read on GitHub
71
39
15
6
4
Would you bet a product on this? Bounded 0–100 and slow moving.
matched fp:700ac234ba899476, topic:rag, topic:retrieval-augmented-generation, readme:retrieval-augmented generation
matched fp:700ac234ba899476, topic:vision-language-model, topic:document-understanding, readme:vision-language