Evolva 是一个工业级本地 Agent 工程框架，面向 CLI/TUI 对话场景，内置规划、工具调用、长短期记忆、技能沉淀、MCP 外部工具生态、Workflow 编排、Trace 可观测性、Eval 评测闭环、Guardrails 策略防护与自我进化能力。

Code of the paper "EvolveNav: Empowering LLM-Based Vision-Language Navigation via Self-Improving Embodied Reasoning" (TPAMI 2026)

Implementation of all RL algorithms in a simpler way

Ghostty-based macOS terminal with vertical tabs and notifications for AI coding agents

LMCache: Supercharge Your LLM with the Fastest KV Cache Layer

Orchestrate multiple coding agents from desktop and mobile

This project aims to replicate mainstream open-source model architectures with limited computational resources, implementing mini models with 100-200M parameters.

大模型技术30讲（原版），中文版、中文批注，30 Essential Questions and Answers on Machine Learning and AI，百面大模型

Handwriting-Transformers (ICCV21)

AI Infrastructure Engineer Learning Track - Production ML infrastructure curriculum (2-4 years experience)

# 微信小程序虚拟支付 · 开发者接入 Skill > **这是什么？** > 这是一份 AI 助手的「技能说明书」。把它添加到你的 AI 编程助手（CodeBuddy、Cursor、Claude 等），AI 就能手把手引导你完成微信小程序虚拟支付的接入。 --- ## 技能目标 为开发者提供一套可直接执行的微信小程序虚拟支付接入流程，帮助快速完成从配置到上线的闭环。 **覆盖能力**： -…

minimal-cost for training 0.5B R1-Zero

使用深度学习方法解析问题 知识图谱存储 查询知识点 基于医疗垂直领域的对话系统

🎨 Local-first, open-source Claude Design alternative. 🖥️ Native desktop app. ⚡ 259+ Skills · ✨ 142+ Design Systems 🖼️ Web · desktop · mobile prototypes · slides · images · videos · HyperFrames 📦 Sa…

Plan and autonomously build a software task end-to-end. One ready-to-paste /goal, adaptive phase count, memory preload + writeback, 3-strike self-healing recovery. Works on Claude Code and Codex.

A curated list of reinforcement learning with verifiable rewards (continually updated)

The absolute trainer to light up AI agents.

Real time interactive streaming digital human

A ground-up LLM engineering project: tokenizer → architecture → training → scaling laws → inference. Starts at 80M, engineered to scale into 1B+ models with minimal changes. Clean, research-ready c…

🌟 推理王国：关于 AI 推理机制的思想实验手册。从信息论、符号逻辑与表示学习出发，系统剖析大模型“智能”的本质。

Agent Reinforcement Trainer: train multi-step agents for real-world tasks using GRPO. Give your agents on-the-job training. Reinforcement learning for Qwen3.6, GPT-OSS, Llama, and more!

SkillOpt is a text-space optimizer that trains reusable natural-language skills for frozen LLM agents through trajectory-driven edits, validation-gated updates, and deployable best_skill.md artifacts.

Multimodal RL training framework for diffusion & omni models

Build a modern LLM from scratch. Every line commented. Explained like we are five.

Getting Started With Chrome DevTools Protocol

An Open Phone Agent Model & Framework. Unlocking the AI Phone for Everyone