FAQs

Answers to the questions people most often ask before relying on Amara for day-to-day terminal work.

Which Model Powers Amara?

Amara runs on the Syntic model, served through api.syntic.ai. The model is tuned for agentic terminal work — reading code, reasoning about command output, and producing safe, reviewable edits. You don’t need to pick or configure a model to get started, though advanced options for choosing models and custom endpoints are available in the Inference section.

Does It Need the Cloud?

Core agent interactions require a connection to the Syntic model endpoint. Features that sync or persist across machines — such as shared conversations and saved memory — activate when connected to your Syntic account. Without an account you can still use Amara locally against the inference endpoint; only the sync and collaboration features are unavailable.

Privacy and Cost

Your prompts and the context Amara reads are sent to the inference endpoint to generate responses. Amara only reads files and runs commands within the scope you allow, and every system-changing action waits for your approval. Usage is tied to your Syntic account and plan; consult your account for current limits and pricing.

Offline Behavior and Feedback

Ordinary terminal use — running commands, viewing output, editing files by hand — works entirely offline. Amara itself needs network access to reach the model. To send feedback, use the in-app feedback option or share your experience through your Syntic account so it reaches the team.