Discover the Best AI Resources

LightX2V

Runs text-to-video, image-to-video, text-to-image, and image editing inference with acceleration, offloading, quantization, and distributed execution for large visual generation models.

Extends MCP so tools can return interactive UIs — charts, forms, dashboards — that render inline in chat clients via sandboxed iframes. Tools declare ui:// HTML resources; an SDK binds them to React, Vue, Svelte, Preact, or Solid.

Deep Agents

Wraps LangGraph in an opinionated harness giving an agent planning, file read/write, sub-agent delegation, and persistent memory out of the box. Aimed at long-horizon work where plain ReAct loops exhaust their context window.

Maestro

Runs several AI coding agents — Claude Code, Codex, OpenCode — in parallel from one keyboard-driven desktop app, each on its own Git worktree. Playbooks batch markdown checklists, group chat coordinates agents, and a web server enables remote control.

Self-hosted personal AI assistant reachable through 20+ messaging channels you already use — WhatsApp, Telegram, Slack, iMessage and more. A local gateway acts as the control plane, stays model-agnostic, and sandboxes tool runs via Docker or SSH.

Hugging Face Skills

Packages Hugging Face ML tasks—dataset creation, model training, evaluation, Hub ops, Spaces deployment—as portable Agent Skills. Each is a SKILL.md folder agents load on demand, running unchanged across Claude Code, Codex, Gemini CLI, and Cursor.

PersonaLive

Generates real-time, infinite-length portrait video from one reference image on a 12GB GPU. Combines implicit facial signals and 3D keypoints with step-distilled diffusion and autoregressive micro-chunk streaming for low-latency live use.

data.gouv.fr MCP Server

Official MCP server for data.gouv.fr, France's national open-data portal: lets chatbots search datasets, query CSV/XLS via the Tabular API, and browse cataloged third-party APIs, all read-only over a public HTTP endpoint that needs no key.

Catalogs reusable Agent Skills for Codex — folders of instructions, scripts, and resources an agent loads to perform specific tasks. Tiers: .system (ships with Codex), .curated, and .experimental. Now deprecated in favor of OpenAI plugins.

TRELLIS.2 — Native and Compact Structured Latents for 3D Generation

Converts images (and other conditions) into high-fidelity, fully textured 3D assets using a 4B-parameter generative model and a field‑free sparse voxel format (O‑Voxel). Handles arbitrary topology, PBR materials, and near real-time mesh/voxel conversions; requires Linux and an NVIDIA GPU with >=24GB memory.

Mental Association Dataset (Rapidata/psychology-association-kiki-bouba-etc)

Collects ~200,000 human responses to 20 visual/semantic association questions (e.g., Bouba–Kiki), with per-response image options and demographic metadata — useful for cross‑cultural perception and evaluation of multimodal systems, but not guaranteed as a rigorously controlled experimental sample.

Forecasts how social scenarios might unfold by running multi-agent simulations: thousands of LLM agents with memory and personalities, seeded from real data, that you steer by injecting variables to 'rehearse the future' in a digital sandbox.