Lists (1)
Stars
Run a documented subset of verl-style OPD on one consumer GPU—typed config, Parquet prompts, and PEFT scale-out artifacts.
Make Every Hardware Intelligent — an open-source Agent OS for hardware intelligence
MemoryBear Equip AI with human-like memory capability
Data-centric LLM training with dynamic sample selection, domain mixture optimization, and example reweighting inside the LLaMA-Factory training loop.
A curriculum-aligned knowledge graph, benchmark, and multimodal training dataset for evaluating and improving curriculum cognition in educational LLMs.
Pack and deploy local AI agents for your team in minutes
本项目实现了从传统机器学习到 BERT 预训练模型的完整迭代，并通过动态量化与知识蒸馏技术，在保障高准确率（93.64%）的同时，实现了极致的推理加速与模型轻量化，完美适配线上高并发推荐场景
Autoharness — a self-learning skill layer for Claude Code — distills skills from your real sessions, updates them as you work, and prunes the ones that stop getting used. No daemon, no benchmark.
Official Implementation of "KungfuBot: Physics-Based Humanoid Whole-Body Control for Learning Highly-Dynamic Skills"
This project is based on "user comments from e-commerce platforms" and uses the "BERT + P-Tuning (soft template)" method to accurately classify the comment texts
Build your own AI agent through gameplay.
Java开源网站内容管理系统(java cms)。使用SpringBoot、MyBatis、Spring Security、Lucene、FreeMarker、TypeScript、Vue3、ElementPlus等技术开发。
Skill-kits is a zero-dependency, single-binary AI Agent Skills management tool for any LLM and multi-agent workflows.
One portable memory layer for every AI agent: local-first, Markdown-native, user-owned, and self-evolving across apps, tools, and workflows.
[ACM MM 2026] Hallo-Live: Real-Time Streaming Joint Audio-Video Avatar Generation
[ICLR2026] This is the first paper to explore how to effectively use R1-like RL for MLLMs and introduce Vision-R1, a reasoning MLLM that leverages cold-start initialization and RL training to incen…
A benchmark for evaluating LLMs on Chinese traditional fortune telling — Bazi (八字) and Ziwei Doushu (紫微斗数).
OpenSquilla — Token-Efficient AI Agent with same budget, higher intelligence density
Molecode presents molecules as code and enables LLMs to operate and reason on chemistry directly.
Automatic and end-to-end scientific research workflow enabled by OpenClaw. Produces state-of-the-art level research results.
OpenClaw / QClaw plugin that visualizes AI agents as 3D NPCs in a game town — with social simulation, a map editor, and a character workshop.
国产开源的桌面级智能体，信创安全，适用于电脑办公、文档处理、IT运维与安全、市场营销等场景。
RoboBrain 2.5: Advanced version of RoboBrain. Depth in Sight, Time in Mind. 🎉🎉🎉
A bio-inspired cognitive memory engine — a new paradigm for Graph RAG.
Manifold learning to detect cell-cell interactions.
A CLI for DaVinci Resolve 18+ — project / media / render / timeline control for humans and AI agents.