About what

All of AI infrastructure

15 layers and 166 categories: everything AI is built from and calls. Search is live on all of them.

Compute Models Inference and serving Training and fine-tuning Data Retrieval Agents Agent browsers and computer use Harness and orchestration Tools and protocols Safety and trust Agent identity, payments and monetisation Dev tools Robotics and physical AI Vertical AI infrastructure

Layer 1 · 14 categories

Compute

Compute is the hardware AI runs on and the companies that rent it out: GPU clouds, AI chips, servers, data centres, and the software that schedules jobs across them. It is where the cost of training and running models is set.

Search liveRecommend: not live yetRoute: not live yetOpen Compute →

AI chips / accelerator clouds AI data centres & colocation AI networking & optical interconnect AI servers & on-prem systems AI storage for GPU clusters Cloud GPU notebooks & dev environments Compute price indexes & GPU price trackers Edge AI accelerators & NPU toolchains GPU cluster schedulers & orchestration GPU marketplaces GPU observability & efficiency Hyperscaler GPU Neoclouds / GPU clouds Serverless GPU

Layer 2 · 24 categories

Models

Models are the trained systems that do the work: language models, and models for speech, images, video, code, translation and forecasting. This layer covers both closed models reached through an API and open models you can download.

Search liveRecommend starts hereRoute: not live yetOpen Models →

Layer 3 · 13 categories

Inference and serving

Inference and serving is everything that turns a trained model into a running service: hosted inference providers, gateways that route between them, serving engines, and the tools that make models smaller and faster.

Search liveRecommend: not live yetRoute: not live yetOpen Inference and serving →

AI gateways & model routers Batch inference Generative media inference platforms Generative media workflow tools Inference benchmarking & load testing Inference kernels & attention libraries Inference providers KV-cache & prompt-caching infrastructure LoRA serving & adapter hubs Model compilers & graph optimizers On-device / edge runtimes Quantization / optimization Serving engines & model servers

Layer 4 · 12 categories

Training and fine-tuning

Training and fine-tuning covers the frameworks, libraries and platforms used to train a model from scratch or adapt an existing one to a task, and the tools that track those experiments.

Search liveRecommend: not live yetRoute: not live yetOpen Training and fine-tuning →

AutoML & hyperparameter optimization Computer vision toolkits & platforms Deep learning & ML frameworks Distributed training frameworks Experiment tracking Feature stores Federated & privacy-preserving learning Fine-tuning platforms ML workflow / MLOps platforms Open-source fine-tuning libraries RL / post-training libraries RL environments & gyms

Layer 5 · 12 categories

Data

Data is what models learn from and read: datasets, labelling services, document parsing, synthetic data, and the pipelines and databases that carry data into AI applications.

Search liveRecommend: not live yetRoute: not live yetOpen Data →

AI training data licensing marketplaces Data & model version control Data catalogs and governance for AI Data pipelines for AI Data quality and curation Datasets Document parsing / OCR Labeling / annotation Multimodal data lakes & AI-native data formats Operational databases for AI apps & agents PII redaction and data privacy for AI Synthetic data

Layer 6 · 11 categories

Retrieval

Retrieval is how an AI system finds the right information at the moment it needs it: web search APIs, vector databases, full-text search engines, crawlers, and the frameworks that tie them into an answer.

Search liveRecommend: not live yetRoute starts here, with search APIsOpen Retrieval →

Chunking and embedding pipelines Enterprise AI search Hybrid and full-text search engines Knowledge graphs RAG frameworks Recommendation & personalization engines Search APIs Text-to-SQL, semantic layers & data agents Vector databases Video understanding & video search APIs Web data / scraping / crawling

Layer 7 · 12 categories

Agents

Agents are AI systems that carry out tasks on their own, and the parts they are built from: agent platforms, memory, voice agents, research agents and the directories where agents and skills are published.

Search liveRecommend: not live yetRoute: not live yetOpen Agents →

Agent UI / frontend frameworks Agent memory Agent, skill & MCP directories Customer support / CX agents Enterprise agent platforms General-purpose autonomous agents No-code agent builders Real-time voice & video agent infrastructure Research / deep-research agents Sales / GTM agents Self-hosted LLM chat interfaces & local LLM apps Voice agents

Layer 8 · 4 categories

Agent browsers and computer use

Agent browsers and computer use covers the tools that let an AI operate a web browser or a whole desktop the way a person would: clicking, typing, reading the screen and completing tasks.

Search liveRecommend: not live yetRoute: not live yetOpen Agent browsers and computer use →

AI-native consumer browsers Browser & computer-use agent SDKs and models Cloud browser infrastructure Computer-use desktop sandboxes

Layer 9 · 4 categories

Harness and orchestration

Harness and orchestration is the software that runs an agent: the frameworks that define its loop and tools, the runtimes that host it, and the engines that keep long tasks going through failures.

Search liveRecommend: not live yetRoute: not live yetOpen Harness and orchestration →

Agent frameworks / SDKs Agent hosting / runtimes Durable execution & workflow engines for agents Structured output & LLM programming libraries

Layer 10 · 6 categories

Tools and protocols

Tools and protocols covers how agents reach the outside world: the standards that connect agents to tools, gateways in front of MCP servers, hosted tool catalogues, and sandboxes where agent code runs safely.

Search liveRecommend: not live yetRoute: not live yetOpen Tools and protocols →

Agent protocols Integrations / iPaaS MCP gateways & proxies Meeting bot & conversation capture APIs Sandboxes / code execution Tool-calling platforms

Layer 11 · 16 categories

Safety and trust

Safety and trust covers the tools that check, test and control AI systems: evaluations, guardrails, security, red-teaming, monitoring, and the platforms that handle governance and compliance.

Search liveRecommend: not live yetRoute: not live yetOpen Safety and trust →

AI agent management platforms AI governance & compliance platforms AI incident & risk databases AI security AI-generated content & deepfake detection Bias, fairness & explainability Confidential AI & private inference Content provenance Evals Guardrails Interpretability & model internals tooling LLM & agent observability Moderation & guard classifiers Prompt management Red-teaming Voice & conversational agent simulation testing

Layer 12 · 5 categories

Agent identity, payments and monetisation

Agent identity, payments and monetisation covers how an agent proves who it is and acts on someone's behalf, how agents pay and get paid, and how AI products meter usage and bill for it.

Search liveRecommend: not live yetRoute: not live yetOpen Agent identity, payments and monetisation →

AI cost management / LLM FinOps Agent auth / OAuth for agents Agent payments / commerce KYA / agent verification Usage metering & billing for AI products

Layer 13 · 9 categories

Dev tools

Dev tools covers the AI tools developers build software with: coding agents, AI code editors, app builders, code review bots, and the tools that make APIs and documentation usable by agents.

Search liveRecommend: not live yetRoute: not live yetOpen Dev tools →

AI app builders (vibe coding)AI coding IDEs AI testing / QA agents API / SDK generators API mocking & testing for agents Agent-native CLIs Agent-ready docs / llms.txt tooling Code review bots Coding agents

Layer 14 · 14 categories

Robotics and physical AI

Robotics and physical AI covers the software and hardware used to build AI that moves in the real world: robot models, simulators, drone and driving stacks, robot platforms with developer kits, and the data they learn from.

Search liveRecommend: not live yetRoute: not live yetOpen Robotics and physical AI →

Autonomous driving software stacks Drone stacks Embodied-agent benchmarks Humanoid & robot platforms with SDKs LLM-to-robot interfaces & robot agents ROS ecosystem Robot cloud & fleet management Robot foundation models Robot hardware dev kits & compute Robot learning datasets Robot perception, SLAM & motion planning libraries Robotics data logging & visualization Simulation Teleop / robot data collection

Layer 15 · 10 categories

Vertical AI infrastructure

Vertical AI infrastructure covers the models, data and platforms built for one field, such as biology, chemistry, finance, health, law, weather and manufacturing.

Search liveRecommend: not live yetRoute: not live yetOpen Vertical AI infrastructure →

Biology & drug discovery AI infra Chemistry & materials AI infra E-commerce AI infra Education AI infra Finance AI infra Geospatial & earth observation AI infra Health & clinical AI infra Legal AI infra Manufacturing & industrial AI infra Weather & climate AI infra

The map is our census of 2026-11-08: 4,643 products across 166 categories.

All of AI infrastructure

Compute

Models

Inference and serving

Training and fine-tuning

Data

Retrieval

Agents

Agent browsers and computer use

Harness and orchestration

Tools and protocols

Safety and trust

Agent identity, payments and monetisation

Dev tools

Robotics and physical AI

Vertical AI infrastructure

Ask about AI infrastructure.

All of AI infrastructure

Compute

Models

Inference and serving

Training and fine-tuning

Data

Retrieval

Agents

Agent browsers and computer use

Harness and orchestration

Tools and protocols

Safety and trust

Agent identity, payments and monetisation

Dev tools

Robotics and physical AI

Vertical AI infrastructure

Ask about AI infrastructure.