Contact Us 1-800-596-4880
  1. Homepage
  2. Enhanced MuleSoft Experience
  3. Adding Services to Your Portfolio

  4. Adding Semantic Services

Adding Semantic Services

Semantic services improve how the system interprets and routes LLM-driven requests across your portfolio. They help teams apply context-aware behavior so the system directs requests to the most relevant tools, services, or pathways that your organization configured.

Use this capability when your workflows need more than static routing. With semantic context in place, the system can support better request matching, more consistent response behavior, and stronger control as usage grows across agents, APIs, MCP servers, and related integrations.

To add semantic services, follow these steps:

  1. Define the semantic service your organization uses for contextual matching.

  2. Associate that semantic capability with the relevant enhanced experience (for example, LLM-related routing paths where supported).

  3. Operate the service as part of your broader governance and monitoring model so behavior remains reliable as traffic and use cases evolve.

Before You Begin

Before getting started, make sure you have:

  • An Anypoint Platform account.

  • Any of these permissions:

    • Exchange: Exchange Contributor

    • Exchange: Exchange Administrator

    • Exchange: Exchange Creator

    For more information, see Enhanced Experience Permissions.

Create a Semantic Service

  1. In the navigation, under Portfolio, expand LLM Proxies, and select Semantic Services.

  2. Select Add Semantic Service.

  3. Select the scale model that matches your traffic profile and operational needs.

    • Basic scale

      Choose Basic scale for faster managed setup. This scale is best for smaller topic sets (up to 6 topics and 10 utterances per topic) and uses managed internal LLM proxy configuration.

    • Advanced scale

      Choose Advanced scale when you need greater scale and external infrastructure control. This scale is best for larger utterance sets per topic. It uses external semantic infrastructure, including an embedding API connection and a vector database connection.

Semantic service setup defines the embedding connection used for contextual matching. Provide values for these fields:

  • Environment

  • Embedding service provider

  • Service label

  • URL

  • Model

  • Auth key

See Also