Agentic AI platforms and enterprise AI agents refer to systems that orchestrate language models, tool calls, memory, and workflows to perform multi‑step tasks autonomously or with human oversight. By 2026 this space spans three overlapping categories: AI Automation Platforms (enterprise orchestration and governance), AI Agent Marketplaces (discovery, templates, and monetization), and Agent Frameworks (developer SDKs and lifecycle tooling). Key offerings illustrate the landscape: LangChain provides an open‑source SDK and commercial platform for building, testing and deploying LLM‑powered agents; Kore.ai targets enterprises with no‑code to pro‑code multi‑agent orchestration plus governance and observability; AgentGPT and Anakin.ai lower the barrier with browser/no‑code agent builders and large libraries of ready apps; GPTConsole and Replit supply developer‑centric SDKs, CLIs and hosted runtimes for production agents; Tabby emphasizes self‑hosted, local‑first coding assistants; GitHub Copilot brings agentic workflows into developer environments; Notion and Coupa Compose show how embedded agents are being productized for knowledge work and procurement. Underpinning these platforms are model families and APIs (OpenAI/Claude/Anthropic/Gemini) whose capabilities, cost models and safety features shape design choices. Current trends include stronger emphasis on governance, observability and lifecycle management; composability across models and tools; marketplaces for reusable agent templates; and enterprise demand for on‑prem/self‑hosted options and data control. Organizations evaluating this category should weigh developer flexibility, no‑code productivity, integration and data governance, plus model portability and cost predictability when selecting platforms or building in‑house agents.