Stars
Completed research on semantic retrieval augmented generation through novel knowledge graph traversal algorithms.
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。
Semi-Structured Agentic Framework. Workflows build themselves as agents discover what needs to be done, not what you predicted upfront.
A custom ComfyUI node for MiniCPM vision-language models, supporting v4, v4.5, and v4 GGUF formats, enabling high-quality image captioning and visual analysis.
Integrating Causal Graphs and Generative AI for Reliable Fact Extraction and Deep Reasoning
The official repo for "VisualWebInstruct: Scaling up Multimodal Instruction Data through Web Search" [EMNLP25]
Convert any MCP server into a Claude Skill with 90% context savings
Codebase for EMNLP 2025 Findings paper "Text or Pixels? Evaluating Efficiency and Understanding of LLMs with Visual Text Inputs"
The official implement of paper 《DaMo: Data Mixing Optimizer in Fine-tuning Multimodal LLMs for Mobile Phone Agents》
not another coding agent, kode is agent cli for everything
Qwen3-VL is the multimodal large language model series developed by Qwen team, Alibaba Cloud.
A quick vibe coded app for deepseek OCR
torchcomms: a modern PyTorch communications API
A lightweight sandboxing tool for enforcing filesystem and network restrictions on arbitrary processes at the OS level, without requiring a container.
Official Implementation of Knowledge Flow Prompting
Convert documentation websites, GitHub repositories, and PDFs into Claude AI skills with automatic conflict detection
CommonForms — open models to auto-detect PDF form fields
Intelligent automation and multi-agent orchestration for Claude Code
The contents of /mnt/skills in Claude's code interpreter environment