An experiment-driven laboratory for understanding LLM sampling strategies. Explore how Temperature, Top-k, and Top-p influence token probabilities and generation diversity through reproducible simulations, visualizations, and real Groq API experiments.

An experiment-driven LLM engineering lab with 9 reproducible Python experiments exploring token economics, API costs, caching, context growth, latency, model routing, rate limits, concurrency, and batching using Groq.

Self-hosted agent runtime: run any harness in isolated sandboxes with streaming UI,HITL, budgets and durable events.

Redacts secrets and personal data in LLM API traffic and restores them in responses

用 MCP 把 Creo 变成 AI 可直接建模的后端：一句话生成可编辑的原生特征树——草图、拉伸、旋转、孔、钣金、装配、工程图、放样、翼型。官方 Toolkit 驱动，每步保存重载回读验证。Windows，Creo 9/10/12。

Provider-neutral telephony adapters, audio pipeline, email capture, STT and RAG for the AI calling agent you already have. TypeScript + Python, zero runtime deps.

你是几B的模型？How many B are you? A meme quiz where you play the LLM, open-sourced with the full creative trail: every prompt, briefs, and a 3-part retro.

Free duck.ai models in Telegram - serverless-first (zero VPS), powered by a pure-JS headless browser that walks through DuckDuckGo's anti-bot challenge.

Several coding agents, one repo, at the same time: a reproducible lab to measure how they coordinate, and Médula, a kernel that decides who waits for whom.

Run Claude Code, Codex and Pi agents toward goals the daemon verifies itself. Give each agent a role and a goal with a completion check; Dudes keeps it working until its own run of the check passes. Rust daemon and CLI, with a 3D office view in the browser.

装在手机里的 AI 互动小说：写下开头、成为主角，历史穿越或自由世界，剧本、挑战、旁观多种玩法。安卓与 Windows，接你自己的 AI 服务（OpenAI 兼容接口）。

An explanation compiler for Claude Code: first-principles explanations in controlled prose, diagrams, interactive pages, or 3Blue1Brown-style videos with local voice, checked for what they leave out and for what you did not need.

Guía gratuita de IA generativa avanzada en español (311 págs.): Transformer, post-entrenamiento, inferencia, GPU, difusión, agentes, seguridad. Free advanced generative AI guide in Spanish. CC0.

MIT-licensed agent framework for the AxisVela ecosystem: governed agent loop, model/tool/memory/event primitives, plugins, sandboxing, and policy.

The open-source brain for personal AI agents. One agent for every user of your app, with memory, schedules, tools and approvals, on a hyperscale serverless architecture where idle agents cost only storage.

Know when an AI conversation thread is still safe to continue — a deterministic, dependency-free continuity planner.

An open-source recreation of AI Co-Scientist: a multi-agent scientific research harness.

A general-purpose LLM harness and coding agent implemented as a single, hackable Bash script.

旁批 · 讲义伴读——选中讲义里的一句话，只用已讲内容讲解；习题只给提示。Windows 便携版

QUASAR: Lowering the Loss Floor of Quantization-Aware Training with Loss-Aware Reconstruction

涛声 TaoSheng：用一部安卓手机全天收音 → 转写 → 认人 → AI 整理成待办与记忆。无需硬件、无需账号，数据只在你手机里。

Qwen Cup multimodal reasoning competition (20/337 → finals 9/20 → defense 3rd 🏆): teacher-distillation SFT + small-dose GRPO RL on Qwen3.5-0.8B, reward-eval verbatim alignment, full data engineering & ablation chain

🔥 Run large open language models on your phone. A C++20 GGUF inference engine that streams models larger than RAM from flash, paces itself by the device's thermal forecast, and powers a private on-device assistant. Android · iOS · macOS · Linux.

Run open AI models on your own computer — a friendly desktop app and local gateway for llama.cpp and more. Private, free, open source.

Student materials for "From NLP to LLM: The Evolution of Language Intelligence" by Mahmmoud A. Mahdi: the book, student workbook, lab manual, and the Sanad project checkpoints.

Agent Skill + Playwright/axe-core verification loop that teaches AI coding agents to build accessible frontends on the first pass. Companion kit for the a11yTO Conf 2026 talk "The Aesthetic-A11y Paradox".

Evidence-backed use cases, prompts, integrations, evaluations, and safety notes for OpenAI GPT-6 Astra.

Every answer leaves a trace back to its page. An MCP server that lets AI agents read long PDFs by address and cite the pages.

Carefully human hand written llm jailbreaks for the latest models that are available on the market. GPT-6 Astra, GPT-6.1 SoL, Opus 4.7, Opus 4.6, Sonnet 4.6, Haiku 4.5, Deepseek V4, Deepseek Flash, DeepSeek V4.1 Pro / Flash, Qwen 3.8 MAX, Qwen 3.8 Omni-Flash, Qwen 3.7 etc

Deep, self-verifying React Native & Expo engineering skills for AI coding agents — 79 skills with checkers that prove the work: architecture, Expo Router, Zustand, TanStack Query, testing, i18n/RTL, accessibility, EAS releases & more. Agent Skills open standard; works with Claude Code, Cursor, Codex & 70+ agents.