⚡️ LLaVA-NeXT: A Strong Zero-shot Video Understanding Model

LLaVA-NeXT is one of the most powerful open source video understanding models available today, superior to all existing VLMs.

New SoTA!

LLaVA-Next demonstrates high data transfer rates and demonstrates excellent results in zero-shot training.

git clone https://github.com/LLaVA-VL/LLaVA-NeXT

▪ Github: https://github.com/LLaVA-VL/LLaVA-NeXT

▪ Blog: https://llava-vl.github.io/blog/2024-04-30-llava-next-video/

▪ HF: https://huggingface.co/shi-labs/CuMo-mistral-7b

▪ Dataset: https://huggingface.co/datasets/shi-labs/CuMo_dataset

https://news.1rj.ru/str/DataScienceT ✅️

MS MARCO Web Search: a Large-scale Information-rich Web Dataset with Millions of Real Click Labels

🖥 Github: https://github.com/microsoft/ms-marco-web-search

📕 Paper: https://arxiv.org/abs/2405.07526v1

✅️ https://news.1rj.ru/str/DataScienceT ✅️

🌟 Step-by-step implementation of the Transformer architecture

This laptop describes in as much detail as possible each step of implementing a transformer from scratch, with the necessary theoretical minimum
For complete enlightenment, you can combine it with video 3b1b

▶️ Jupyter Notebook

✅️ http://t.me/codeprogrammer ✅️

🌟 Verba is an open-source application to provide an end-to-end, optimized and user-friendly interface for Retrieval-Augmented Generation

- pip install goldenverba

Verba is a fully customizable AI assistant for querying and interacting with your data, both on-premise and deployed in the cloud.
Answers questions related to your documents, obtains information from existing knowledge bases. Verba combines state-of-the-art RAG technologies with Weaviate's context-aware database. Choose between different RAG frameworks, data types, partitioning and search methods, and LLM providers depending on your specific use case.

Allows you to use different LLMs: both from HuggingFace and Ollama, and from OpenAI, Cohere and Google.

🖥 GitHub