Providers and models | ATX AnalyticsTable Docs

Providers and modelsThe six model adapters, endpoint classes, the default model, the per-session model picker, where keys are kept, and models installed by your organisation. ATX AnalyticsTable does not ship a language model. Ask Greg, the AI_PROMPT formula and the pipeline's AI analysis step all talk to a model that you, or your organisation, configure as a provider profile: an adapter, an endpoint, a model name and an endpoint class. Until at least one is configured, the Ask Greg panel says "No model is configured" and asks you to add one under Providers.

Where to find it

Ask Greg (beta) → Models → Providers

What this page covers

Adapters
Provider fields
Endpoint classes
The default model and the per-session picker
Where keys are kept
Models installed by your organisation
Group Policy controls

Settings and controls documented

openai-compatible
azure-openai
anthropic
google-gemini
completion-only
custom-http
https://<resource>.openai.azure.com
https://<resource>.services.ai.azure.com/models
https://<resource>.openai.azure.com/openai/v1
Name
Adapter
Endpoint
Model or deployment
Embedding model
API key
Used for
Endpoint class
Default for answers
Enabled
on-prem
private-cloud
public-cloud
Windows
macOS
Linux
apiKeyCipher
apiKeyEnv
apiKeyFile
apiKey
(none)
Turn off AI features
Restrict AI providers
Managed AI policy file

Full page

The full page — procedures and the explanation of every setting — is available to signed-in users.

ATX Institute on LinkedIn

This site needs JavaScript for the full experience. Every page above is reachable without it.