# Catalogue > Public catalogue components. JSON: https://leyline.ahensya.com/api/v1/components - [Model Context Protocol](https://leyline.ahensya.com/components/model-context-protocol.md): Open protocol so tools, data, and prompts plug into many harnesses. The connect slice Leyline ranks, not a marketplace. - [Alibaba Cloud Model Studio](https://leyline.ahensya.com/components/alibaba-model-studio.md): Model APIs for text, vision, speech and embeddings, with compatible developer interfaces. - [BytePlus ModelArk](https://leyline.ahensya.com/components/byteplus-modelark.md): Model APIs and SDKs for reasoning, multimodal understanding, and image and video generation or editing. - [TypeSafe Jev](https://leyline.ahensya.com/components/typesafe-jev.md): Returns typed judgments and probabilities from software state and narrowly defined questions. - [OpenAI GPT Image](https://leyline.ahensya.com/components/openai-gpt-image.md): Image-generation and editing models available through OpenAI developer APIs. - [Laya](https://leyline.ahensya.com/components/laya.md): Laya = Apache-2.0, non-autoregressive System One decision model family: given a state (text / email / ticket / JSON) and typed questions (choice / score / noul), returns typed answers with probabilities in one forward pass. Trained with RLCD (reinforcement learning against strictly proper scoring rules). - [Needle 3](https://leyline.ahensya.com/components/needle-3.md): Needle 3 is Cactus Compute’s on-device “automation foundation model”: a laddered / sliceable tiny model (post: 8–29 MB CQ2-bit binary; 25–121M params; depths 2–20 layers from one weight set) built on their Simple Attention Networks, aimed at tool calls, structured extraction, and local text embeddings — not open-ended chat. - [OpenRouter](https://leyline.ahensya.com/components/openrouter.md): Unified model and inference router. Listed honestly as an inference gateway — inspiration-adjacent, not Leyline's core. - [MCP / AI Gateway](https://leyline.ahensya.com/components/mcp-ai-gateway.md): Gateway layer that routes model calls, discovers MCP tools, and applies identity, policy, and spend controls before invocations. - [Toloka](https://leyline.ahensya.com/components/toloka.md): Human-in-the-loop training data, RL environments, and evaluation for AI agents and LLMs. - [BlockRun](https://leyline.ahensya.com/components/blockrun.md): Provides model and service access through documented APIs, SDKs and MCP tools. - [Concentrate.ai](https://leyline.ahensya.com/components/concentrate-ai.md): Concentrate is a managed LLM gateway: one API base URL (https://api.concentrate.ai/v1) for hundreds of models across all major providers (OpenAI, Anthropic, Gemini, DeepSeek, etc.). You pick the model and provider; Concentrate runs the request, applies your fallbacks and limits, and surfaces spend/logs. - [OpenRouter](https://leyline.ahensya.com/components/openrouter-c833b6.md): OpenRouter is a unified large language model inference router and model marketplace: one OpenAI-compatible HTTP application programming interface across many underlying inference providers, with automatic fallbacks and cost-aware routing. The marketplace object is models and inference providers (tokens in → tokens out), not installable agent tools ranked by dual human-and-agent reputation. - [Prime Agent](https://leyline.ahensya.com/components/prime-agent.md): Prime Agent is an MIT-licensed open-source coding and research agent from Prime Intellect, built around a Recursive Language Model (RLM) runtime and a persistent IPython control environment. The model works through programmatic Python: file/shell operations, data transforms, skills, context management, and recursive subagents are all callable from the durable kernel rather than exposed as a large fixed tool palette. - [PulseMCP](https://leyline.ahensya.com/components/pulsemcp.md): Directory of MCP servers, clients, use cases, and tools plus an MCP newsletter, with listing ingestion currently paused for process rework. - [telemetry.dev](https://leyline.ahensya.com/components/telemetry-dev.md): Hosted observability for AI/LLM apps. Product line: every model call, tool step, and retrieval in one OpenTelemetry trace, with tokens, cost, latency, and errors. Ingest via TypeScript or Python SDKs, or any OTLP exporter. Cost is computed server-side against 4700+ models. - [Deep Agents](https://leyline.ahensya.com/components/deep-agents.md): Deep Agents (langchain-ai/deepagents) is LangChain AI’s batteries-included agent harness: production-ready, extensible, with filesystem, context management, sub-agent delegation, skills, and long-term memory (PyPI summary). README principles: opinionated, extensible, model-agnostic, production-ready on LangGraph + LangSmith. - [Invisible Playwright MCP](https://leyline.ahensya.com/components/invisible-playwright-mcp.md): A locally run MCP browser agent and optional web UI that automates a stealth Firefox browser for web browsing, scraping, and page interaction. - [kogiQA MCP Web Browser](https://leyline.ahensya.com/components/kogiqa-mcp-web-browser.md): A Model Context Protocol server that gives LLM agents selector-free browser automation for debugging, documenting, and testing web applications through natural-language instructions. - [MiniMax](https://leyline.ahensya.com/components/minimax.md): Developer APIs for language and multimodal generation, including speech, images, video and music. - [Open Notebook](https://leyline.ahensya.com/components/open-notebook.md): Open Notebook is an open-source implementation of Google NotebookLM-style workflows with more model choice and deployment flexibility. The project positions itself as a private, multi-model, local/self-hostable notebook for ingesting sources, searching them, chatting with grounded context, and generating podcasts from research material. - [QuiverAI](https://leyline.ahensya.com/components/quiverai.md): QuiverAI ships foundational models for generating, editing, and animating editable vector/SVG assets (logos, illustrations, diagrams, line work; typography & animations marked coming soon on marketing site). Arrow 2 is the Sep 2026 model line called out as higher quality and faster, in App and API. - [41 Open-Source LLMs Evaluated on 19 Benchmarks](https://leyline.ahensya.com/components/41-open-source-llms-evaluated-on-19-benchmarks.md): A repository of local benchmark evaluations comparing 41 open-source large language models across 19 tasks, with rankings, scripts, notebooks, results, and utilization logs. - [Factory Droid](https://leyline.ahensya.com/components/factory-droid.md): A coding-agent terminal interface for scoped repository changes and reviewable development tasks.