Discover the Best AI Resources

ANE Training — Backpropagation on Apple Neural Engine

Runs transformer forward+backward training directly on Apple's Neural Engine by reverse-engineering private ANE APIs; includes per-layer ANE kernels, INT8 optimizations and benchmarks. Proof-of-concept only—relies on undocumented APIs, has low utilization and CPU fallbacks.

Obsidian Mind

Provides AI coding agents with persistent memory inside an Obsidian vault—preserving session context, decisions, and notes across sessions. Integrates hooks/commands for Claude Code, Codex CLI, and Gemini CLI and optionally uses QMD for semantic recall; aimed at developer workflows.

PM Skills Marketplace

Encodes product-management frameworks into reusable AI skills and slash-commands — 100+ skills across 9 plugins and chained workflows (e.g., /discover, /write-prd) that integrate with Claude Code/Cowork and other assistants to guide discovery, strategy, execution, and shipping.

Paperclip

Orchestrates teams of AI agents to pursue company goals: hire agents, assign tasks, enforce budgets, and audit work from a single dashboard. Combines org charts, persistent agent state, heartbeat scheduling, approval gates and cost controls for long-running autonomous workflows.

Council of High Intelligence

Convenes 18 deliberately polarized AI personas to produce structured, multi-round deliberations on hard questions across multiple LLM providers. Key features: multi-provider auto-routing, enforced dissent/novelty rules, triad/panel modes and CLI integration for Claude Code/Codex. Good for high-stakes product, strategy, or safety decisions.

google/gemma-4-E4B-it

An instruction‑tuned Gemma 4 E4B multimodal model on Hugging Face that accepts text, images and audio and generates text; notable for 128K long context support, built-in thinking mode, and an on‑device‑friendly E4B architecture under an Apache‑2.0 license.

Agent Governance Toolkit

Performs deterministic, sub-millisecond policy checks on every agent action (allow/deny + audit) and adds zero-trust identity, execution sandboxing, and SRE features. Covers the OWASP Agentic Top 10 and is designed to sit between agent frameworks and runtime actions for auditable, low-latency governance in production.

OBLITERATUS

Identifies and surgically removes the internal activation directions that trigger refusal behavior in large language models, with one-click options on a HuggingFace Space or a local Python API. Combines multiple extraction methods (SVD, whitened SVD, sparse autoencoders), reversible steering, and analysis-informed verification to quantify capability and refusal trade-offs.

whichllm

Detects your GPU/CPU/RAM and ranks HuggingFace models that will actually run and perform best on your machine, using live, recency-aware benchmarks and a one-line CLI for simulation, planning, and instant local runs.

OpenAI Plugins

A curated collection of Codex plugin examples demonstrating plugin manifests, companion surfaces (skills, hooks, assets), and sample integrations. Highlights richer, opinionated examples for Figma, Notion, iOS/macOS/web builds and MCP-backed bundles — useful for prototyping assistant plugins.

WorldWideView

Visualizes live global data on a CesiumJS 3D globe via a plugin-driven pipeline and real-time WebSocket DataBus. Supports dynamic plugin marketplace, an opt-in Agent Bus for external LLM/MCP control, and self-hosting with Docker and PostgreSQL.

Qwen3.5-9B-Claude-4.6-HighIQ-THINKING-HERETIC-UNCENSORED

Community fine-tuned multimodal Qwen3.5-9B using Claude 4.6 distilled data to change the model's 'thinking' behavior; offers an uncensored 'heretic' flavor with image-text-to-text I/O, benchmark comparisons, and deployment notes for inference frameworks.