—
—
PROWBench: evaluating whether video models faithfully render what a program specifies.
—
AI API gateway / relay: 307 models behind one OpenAI-compatible key. image2.5 $0.0085/image, Seedance 2.0 Mini $0.01056/sec, LLM from $0.0228/M tokens. USD settlement, pay-as-you-go, $1 minimum top-up.
Official implementation of paper PluginRSI: Recursive Improvement of Agent Harnesses with Reusable Plugins
Production-grade SEO + GEO content factory for Claude Code — research, write, fact-check, optimize, publish to WordPress, monitor. Battle-tested by Loamwright 沃匠 SEO agency.
A curated collection of papers, research blogs, and open-source projects on recursive self-improvement (RSI), self-modifying agents, and iterative self-training.
The highly anticipated open-source repository for JEV as Policy enables one-click setup of the simulation environment. Evaluations of Astra + JEV on benchmarks such as RoboTwin will also be released soon.
Bonsai-27B x NInfer local deployment project (RTX 5080, ternary quantization)
Scroll-driven websites from one YAML file: generated film, live 3D, data stories, maps. Runs locally, no API keys; Claude can drive it via MCP.
A Closed-Loop Benchmark for Long-Horizon Real-World Driving with Explicit Navigation Routes
AI API gateway / relay: 307 models behind one OpenAI-compatible key. image2.5 $0.0085/image, Seedance 2.0 Mini $0.01056/sec, LLM from $0.0228/M tokens. USD settlement, pay-as-you-go, $1 minimum top-up.
AI API gateway / relay: 307 models behind one OpenAI-compatible key. image2.5 $0.0085/image, Seedance 2.0 Mini $0.01056/sec, LLM from $0.0228/M tokens. USD settlement, pay-as-you-go, $1 minimum top-up.
AI API gateway / relay: 307 models behind one OpenAI-compatible key. image2.5 $0.0085/image, Seedance 2.0 Mini $0.01056/sec, LLM from $0.0228/M tokens. USD settlement, pay-as-you-go, $1 minimum top-up.
—
Shopping agent research project
—
AiNxt Enterprise is an open-source enterprise AI platform designed to help organizations build, deploy and manage AI applications, agents and workflows. It brings AI models, tools, automation and governance together in a single platform, with the flexibility to run on an organization’s own infrastructure.
MiniMax-H3 Fun ControlNet for ComfyUI. Depth, canny, pose, HED or MLSD video driving H3 frame by frame, built on ComfyUI's own H3 blocks.
—
Typed judgment tools for MCP agents. TypeSafe's Jev model as verify, screen, find, classify, rerank, decide, compare, extract, review, gate, and score: the model judges, policy decides auto, review, or escalate.
Tools for CES abuse
Open-source Quest 3 PCVR streamer (ALVR fork) with PyroWave wavelet video decoded in Vulkan on the Quest's GPU. Quality 120 Hz over USB out of the box; tested profiles picked in the headset.
AI API gateway / relay: 307 models behind one OpenAI-compatible key. image2.5 $0.0085/image, Seedance 2.0 Mini $0.01056/sec, LLM from $0.0228/M tokens. USD settlement, pay-as-you-go, $1 minimum top-up.
AI API gateway / relay: 307 models behind one OpenAI-compatible key. image2.5 $0.0085/image, Seedance 2.0 Mini $0.01056/sec, LLM from $0.0228/M tokens. USD settlement, pay-as-you-go, $1 minimum top-up.
An open-source, hands-on mobile robotics course covering Jetson, ROS 2, AI perception, SLAM, autonomous navigation, semantic understanding, and robotic manipulation.
—
Lightweight, GPU-accelerated reinforcement learning for Isaac Lab, mjlab, and Gymnasium.
Windows 網路流量轉發與代理工具:WinDivert 驅動層封包攔截、多進程攔截規則、多網卡路由與故障轉移、IPv6 雙棧、SOCKS5/HTTP 代理、PySide6 即時監控介面 (MIT)
Agent Skills for working UX designers. Challenge ideas, find blind spots, use the real design system, preserve evidence and decisions, and carry UX intent into engineering.
Tamper-resistant regression gate for AI coding-agent instructions—AGENTS.md, CLAUDE.md, Cursor, Copilot, Gemini, and more.
微信公众号文章实时排版 Skill:5 套主题、组件设计、联动换色、富文本复制与风格学习
AI Slop 탈곡기: 이 글은 흥미롭지 않습니다. 한국어 글의 AI 말투와 과잉 설명을 다듬는 Codex 스킬.
Not a Mused
DolphinBench - Mapping the Pareto frontier of Agent Memory
Interactive web app for exploring, visualizing, filtering, and exporting CSV data — built with Python and Streamlit.
[Emergent Multi-View Geometry Through Self-Distillation]: 3D SSL through multi-view self-distillation, called Poincar3
reverse-engineered FLUX Kontext Max API: 1K $0.064. model ID `flux-kontext-max`, OpenAI-compatible, async task + cost in every response, $1 minimum top-up. flux-kontext-ext-reverse-api.
AI API gateway / relay: 307 models behind one OpenAI-compatible key. image2.5 $0.0085/image, Seedance 2.0 Mini $0.01056/sec, LLM from $0.0228/M tokens. USD settlement, pay-as-you-go, $1 minimum top-up.
AI API gateway / relay: 307 models behind one OpenAI-compatible key. image2.5 $0.0085/image, Seedance 2.0 Mini $0.01056/sec, LLM from $0.0228/M tokens. USD settlement, pay-as-you-go, $1 minimum top-up.
AI API gateway / relay: 307 models behind one OpenAI-compatible key. image2.5 $0.0085/image, Seedance 2.0 Mini $0.01056/sec, LLM from $0.0228/M tokens. USD settlement, pay-as-you-go, $1 minimum top-up.
面向中文研究生与科研工作者的学术辅导 skill 套件(支持 Claude Code / Codex 等编程智能体):选题 → 检索 → 成文 → 评审 → 投稿全链路 25 个命令,引用核验以真实 API 响应为准、不凭模型记忆编造,AI 管效率、人管研究决策——辅导不代笔
一个框架化蒸馏作家的skill
GenRouter & GenCanvas
EMH: a diagnostic Hermes Agent profile (unofficial)
Official code for "GROW: Group-Relative Advantage-Weighted On-Policy Reinforcement Learning of Autoregressive-Diffusion Text-to-Speech Model"
「video-catcher」面向 Codex、Claude Code 等本地 Agent 的视频资源发现与下载 Skill。
Your coding agents remember what you decided — and what you changed. Plain-Markdown memory in Obsidian, shared by Claude Code, Codex and Cursor over MCP. Every answer cites its page; when the vault does not know, it says so. Imports your past sessions on install.
SGLang for GLM-5.3-Flash on two or four RTX PRO 6000 Blackwell GPUs (SM120), with TP2 W4A16 and TP4 NVIDIA NVFP4 profiles.
在codex里调用中转站的API进行image2生图,适用于使用中转站但没有gpt会员的宝子们,唯一需要的是中转站的URL和你用的中转站image2的API KEY。我的抖音:萌猪过河
—
Photographs to a textured mesh on one 8 GB NVIDIA card, offline. Built on TRELLIS.2, with a four view path, and the photographs' own pixels projected onto the texture so printed text stays readable.
Measure your speakers with a microphone and generate a validated, protective parametric calibration for the Omarchy bar.
—
灯暗一点,漫画翻慢一点。支持页内悄悄话、点赞分享与按需看图的独立共读放映厅。
智能通用视频粗剪与口误废案剔除 Skill:自动检测语音停顿与空白,结合大模型并发滑窗识别口误重讲与废案,硬件微淡化导出无爆音成片。
MC-Sparse: Deconstructing and Closing the Dense–Sparse Attention Gap in Diffusion Transformers.
AI API gateway / relay: 307 models behind one OpenAI-compatible key. image2.5 $0.0085/image, Seedance 2.0 Mini $0.01056/sec, LLM from $0.0228/M tokens. USD settlement, pay-as-you-go, $1 minimum top-up.
Embodied Passive Aeroacoustic Perception Enables Relative Sensing and Pursuit Between Aerial Robots