Un-LOCC: Universal Lossy Optical Context Compression for Vision-Based Language Models Achieve nearly 3x token compression at over 93% retrieval accuracy using existing Vision-Language Models.

🐊 Snappy: A vision-first document retrieval framework, using ColPali + DeepSeek OCR. Like the project? Drop a star! ⭐

A Reproduction of GDM's Nested Learning Paper

A Comprehensive Survey on Knowledge Distillation

PowerRAG: Build smarter, more accurate RAG systems with OceanBase's multi-modal retrieval.

The AI-Native Search Database. Unifies vector, text, structured and semi-structured data in a single engine, enabling hybrid search and in-database AI workflows.

Democratizing AI scientists with ToolUniverse

A Powerful Retrieval Augmented Generation Tool for MongoDB Vector Search

paper list, dataset, and tools for radiology report generation

The Intelligent Inference Scheduler for Large-scale Inference Services.

The #1 open-source SWE-bench Verified implementation

Build AI agents with Go. Multiple providers, multiple models, one API 🧙

微舆：人人可用的多Agent舆情分析助手，打破信息茧房，还原舆情原貌，预测未来走向，辅助决策！从0实现，不依赖任何框架。

Official implementation of "Continuous Autoregressive Language Models"

NOFX: Defining the Next-Generation AI Trading Operating System. A multi-exchange Al trading platform(Binance/Hyperliquid/Aster) with multi-Ai competition(deepseek/qwen/claude)self-evolution, and re…

Pure-Python library for computing fuzzy hashes (ssdeep)

Interface for AutoDock, molecule parameterization

Rust multi‑backend OCR/VLM engine (DeepSeek‑OCR, PaddleOCR‑VL, DotsOCR) with DSQ quantization and an OpenAI‑compatible server & CLI – run locally without Python.

A Genomic Foundation Model

Implementation of ConvLSTM to predict the binding affinity of protein-ligand complexes from molecular dynamics simulations

PyTorch implementation of Masked Autoencoder

基于 IMO25 和 Deep Research 的 Deep Think Agent

An AI agent system for solving International Mathematical Olympiad (IMO) problems using Google's Gemini, OpenAI, and XAI APIs.

Python binding for curl-impersonate fork via cffi. A http client that can impersonate browser tls/ja3/http2 fingerprints.

A Rust machine learning framework.

Open-Source Low-Latency Accelerated Linux WebRTC HTML5 Remote Desktop Streaming Platform for Self-Hosting, Containers, Kubernetes, or Cloud/HPC

ImageMol is a molecular image-based pre-training deep learning framework for computational drug discovery.