Large language models on the ServiceNow AI Platform
ServiceNow AI Platform® utilises large language models (LLM) for generative AI and agentic AI functionality. You can choose the ServiceNow® LLM or supported third-party models. LLMs bring about the Generative AI application capabilities by understanding and generating human language, and processing vast amount of data.
Third-party model strategy
| Model provider | Large language model | Version | Deployment policy |
|---|---|---|---|
| ServiceNow® | Now LLM Service | Latest update for your instance | Default |
| Microsoft Azure | Azure OpenAI | OpenAI GPT-4.1 and GPT-4.1-mini | https://learn.microsoft.com/en-us/azure/ai-services/openai/how-to/deployment-types#global-standard |
| Google Gemini | Gemini 2.5 Flash and 2.5 Pro | https://cloud.google.com/vertex-ai/generative-ai/docs/learn/locations | |
| AWS | AWS Claude | Claude 3.7 Sonnet | https://docs.aws.amazon.com/bedrock/latest/userguide/models-regions.html |
Large language model selection
This selection is supported at various levels such as, skill, skill group and instance levels. However, the controls that define LLMs mapping to different skills, allowed at various global locations, are configured and approved in AI Control Tower by the AI steward. The model provider options for custom skills created in Now Assist skill kit and AI Agent Studio are also configured in AI Control Tower.
- Edit the model provider at the instance, skill group and skill levels
- Deactivate active skills which are non-compliant with fallback as No. See Explore AI model providers to learn more.
- View audit history on policy updates by AI steward in AI Control Tower
- Select allowed model providers across domain separated instances.
See Manage large language models to learn about managing.
For more information on selecting and updating model providers, see Edit model providers.