MaxAI Supported LLMs

HCL MaxAI supports both cloud-hosted and on-premises LLM deployments.

Cloud / Hosted LLM

Item	Cloud Hosted LLM	On-Prem / Local Hosted LLM
Supported Model	OpenAI GPT-5.2	GPT-OSS 120B
Usage	Default model used across MaxAI features	Primary local model
Endpoint Type	Managed LLM Endpoint	Local Inference Endpoint; Inference Engine: vLLM
GPU Requirement	Not Required	Required; Recommended NVIDIA A100 / H100 Class GPUs
Infrastructure Ownership	OpenAI / Hosted Service	Customer Managed
Deployment Model	Cloud	On-Premises