GREATMEMORY.

// Deployment options

Simple cloud marketplace or private cloud.

GreatMemory currently runs one controlled workspace per install, with multi-workspace access planned for teams that need one user to move between multiple data estates. One codebase ships as marketplace packaging or a private-cloud install; behavior stays the same.

Simple cloud path

Cloud marketplace

Install GreatMemory from cloud marketplace packaging into the customer's own AWS, Azure, or Google Cloud account. The customer keeps the account, network, database, logs, and LLM keys.

  • Single workspace today; multi-workspace access is on the roadmap
  • Marketplace entitlement at deploy time
  • BYO public LLM providers or local providers
  • Cloud-native logging and optional bundled Grafana
  • Starter UI, REST, streaming chat, ETL, and MCP surfaces

Custom service path

Private cloud

Run the same stack in your private environment. We can prescribe the services your team should buy and install, or we can procure, install, configure, and maintain the required services for you.

  • Helm, Terraform, or compose packaging
  • Custom quote based on scope and operating model
  • Full maintenance available after install
  • Private telemetry posture by default
  • Optional appliance mode for local-only AI and no external egress

Public cloud topology

GreatMemory deploys from a marketplace listing into your own AWS, Azure, or Google Cloud boundary. You keep the account, network, database, object storage, logs, and LLM provider policy.

no shared public SaaS

GCP

Google Cloud

Marketplace

Google Cloud Marketplace

GreatMemory

governed single workspace

ComputeCloud Run or GKE
DataCloud SQL, AlloyDB, Cloud Storage
LLMVertex AI or external LLM keys
OpsCloud Logging + Monitoring

AWS

Amazon Web Services

Marketplace

AWS Marketplace

GreatMemory

governed single workspace

ComputeECS, EKS, or App Runner
DataRDS/Aurora, S3, optional OpenSearch
LLMBedrock or external LLM keys
OpsCloudWatch + managed metrics

Azure

Microsoft Azure

Marketplace

Azure Marketplace

GreatMemory

governed single workspace

ComputeContainer Apps or AKS
DataAzure PostgreSQL, Blob Storage
LLMAzure OpenAI or external LLM keys
OpsAzure Monitor + Log Analytics
Marketplace entitlement
Customer-owned cloud account
Single workspace today
Cloud-native logs and metrics

// Private cloud

Local LLM, local data, expandable capacity.

Private cloud can stay entirely inside your boundary: local LLM server, local database, local storage, backup path, and users accessing the governed GreatMemory workspace. Capacity can grow from a compact GB10-style appliance to 100GB+ memory servers and multi-GPU nodes.

Private cloud topology

Local inference, local data, backups, and scale-out capacity inside your boundary.

scales inside your boundary
SmallMediumLarge

Capacity expands by adding local GPU, database, storage, and backup resources inside your boundary.

Local LLM

DB + storage

Small

Compact private appliance or GB10-style local box.

  • Local LLM through Ollama, llama.cpp, MLX, or LocalAI
  • SQLite or managed Postgres
  • Lower token throughput, strong data locality
A100
A100
A100
A100

Medium

100GB+ memory local LLM server for higher-throughput search/chat.

  • vLLM or llama.cpp for selected 100 tokens/sec-class workloads
  • Postgres + pgvector, local object storage, scheduled backups
  • Best fit for legal, support, and transcription teams

Large

Scale-out private cloud for heavier concurrency and growth.

  • Multi-GPU servers with vLLM, TensorRT-LLM, or NVIDIA NIM
  • HA database, storage tier, backup vault, and observability
  • Adds capacity as users, data, and models grow

Local throughput depends on the model, quantization, context length, and hardware. We validate the target profile during sizing instead of promising one number for every workload.

// Operating model

Managed by you, maintained by us, or fully managed.

Contact us for a quote based on whether you want advice only, installation and maintenance, or a fully managed private-cloud service.

01

Customer managed

We prescribe the required services and architecture; your team buys, installs, and operates them.

02

Install + maintenance

Your team owns the environment; we install GreatMemory, wire the services, handle upgrades, and maintain the stack.

03

Fully managed by us

We can procure, install, operate, maintain, and support the private-cloud service under a custom quote.

// Scope

What we clarify before quoting.

The deployment is simple when the ownership boundary is clear: who buys the cloud services, who installs them, who maintains them, and which monitoring and LLM providers are allowed.

Target cloud or private environment
Database choice: SQLite, Postgres, or MySQL
Hosted or local LLM provider policy
LDAP/AD, role, and API access needs
Telemetry: Grafana, cloud-native monitoring, or both
Maintenance, upgrade, and support expectations