Keep these considerations in mind when using Agentforce Merchandising for B2B Stores.
Required Editions
Language and Locale Support
This feature supports English in these locales.
| Locale | Code |
|---|---|
| Locale1 | en_LO |
| Locale2 | en_LO |
Large Language Model Support
This feature supports these models. For more information about LLM support for the AI agents platform and reasoning engine, see Consideration for Agents.
| Model |
|---|
| OpenAI GPT-4 Turbo |
Einstein Trust Layer Service Support
This feature supports these Trust Layer services.
| Service | Description |
|---|---|
| Dynamic Grounding with Secure Data Retrieval | Not required |
|
Prompt Defense (System policies and prompt injection detection) |
Supported by the Platform (Employee Agent) |
| Data Masking | Not supported |
| Toxicity Detection | Supported (default from the platform) |
| Audit and Feedback | Supported (default from the platform) |
Region Support
Merchandising Agent doesn’t require geo-aware LLM request routing as described in Geo-Aware LLM Request Routing on the Einstein Generative AI Platform.
Feature and Einstein Requests Usage
Einstein Requests is a usage metric for generative AI. The use of generative AI capabilities in either a production or sandbox environment consumes Einstein Requests and possibly Data 360 credits. To learn more, see Einstein Usage.