Starred repositories
Chronos: Pretrained Models for Time Series Forecasting
A modern multimodal knowledge graph with type-specific metadata across biomedical domains.
Low-code programming for event-driven applications
CompBioBench: A benchmark of 100 diverse, verifiable questions for agents for computational biology (scripts for running agents)
Scalable toolkit for efficient model reinforcement
Visualize live LangGraph execution and see how your agent thinks as it runs.
Automatically detect LangGraph code in Python files and visualize graph structures directly in VS Code. Shows nodes, edges, and connections in a beautiful, theme-integrated webview panel.
TradingAgents: Multi-Agents LLM Financial Trading Framework
A framework for comprehensive diagnosis and optimization of agents using simulated, realistic synthetic interactions
Official code base for LeWorldModel: Stable End-to-End Joint-Embedding Predictive Architecture from Pixels
Offical Implementation for "Recursive Multi-Agent Systems"
LLaDA2.0-Uni: Understanding and Generation the World.
This is the repo for the paper A Self-Evolving Framework for Efficient Terminal Agents via Observational Context Compression
A Foundational Evolving Agent Framework for Agentic Science at Scale.
A command-line interface (CLI) tool for managing files and knowledge bases in OpenWebUI.
Mixed-precision quantization for LLMs. Every layer refracts into a different format based on its sensitivity. Native compressed-tensors export, validated on Qwen3.6-35B-A3B MoE with MTP speculative…
AI enabled pair programmer for Claude, GPT, O Series, Grok, Deepseek, Gemini and 300+ models
TurboQuant+ KV cache compression for vLLM. 3.8x smaller KV cache, same conversation quality. Fused CUDA kernels with automatic PyTorch fallback.
A Python implementation of Google's TurboQuant framework (WIP)
Share a single keyboard and mouse between multiple computers.
MCP server that saves Claude Code tokens by delegating bounded tasks to local or cloud LLMs. Works with LM Studio, Ollama, vLLM, DeepSeek, Groq, Cerebras.
Generate text, images, video, speech, and music by MiniMax.