Resources
Guides for using iMothership at work
Get started, understand the product, and fix things when they go sideways.
Model usage
Models are chosen by the organisation, then used by projects. That is how iMothership keeps AI on track and keeps token spend from wandering.
How approval works
An admin opens Settings, then AI, and sets up the approved AI sources. A project can only select from that list. You can also set models and pricing used in reports for each source.
Providers we configure for business use
iMothership uses business or API services, not consumer chat apps, when it sends work to a model. The current set includes:
- Anthropic
- Google Gemini / Vertex AI
- OpenAI
- Replicate
- Fireworks AI
- Groq
- ElevenLabs
- SpaceX AI / xAI
Where a provider offers no-training or zero-data-retention controls, those are enabled for sensitive data such as Google Workspace content. The living detail is on AI & data safety.
Pick the right model for the step
Use a cheaper, faster model for routine steps. Use a stronger model when the step needs judgement. Match jurisdiction and security needs when data cannot leave a region or a provider.
If chat or a task fails, check that the organisation still has that AI source approved, that the project is allowed to use it, and that the plan covers the run. Then read the log for that action.