FAQ
If you self-host your LLM, prompts, screenshots, and outputs can remain inside your network. With hosted providers, data flows only to the model you connect with, using your credentials; we do not sell or train on your data. (Configure zero-telemetry mode and local log retention if needed.)
It depends on your LLM/provider and hardware. DeepSeek and similar models are optimized for high throughput; ScreenHelper executes UI steps as soon as instructions are ready, often faster than manual use. (api-docs.deepseek.com)
DeepSeek 3.1, Gemini 2/+, GPT, Claude, Grok — and we’ll keep adding more.
Yes, if your connected LLM supports tools/browse with your account, ScreenHelper can ask it to fetch data, cite sources, and continue the on-screen workflow.
No. You can run chat-only, which avoids STT/TTS costs entirely.
With DeepSeek’s current list prices (deepseek-chat), a typical workload of ~200 input + 150 output tokens per second plus 1 screenshot/second is roughly \$1.27–\$2.73 per hour (≈ €1.09–€2.34/hour), depending on how many tokens each screenshot becomes. Actuals vary by provider, time of day, and prompt caching. (api-docs.deepseek.com, European Central Bank)
Windows and macOS at launch (Linux roadmap). (Adjust this line as needed.)
When you buy €600 Lifetime, you receive a license code that permanently unlocks ScreenHelper on one device (per-device binding). No future payments to Avataris for that device.
Yes. It’s purchasable now, but activations may wait in a queue if demand spikes to keep service quality high.
TRY BETA VERSION
Email us at request@avataris.ai to receive a download link. We may release a web version here in the future, so stay tuned!
For updates on new products and demo releases, follow us on LinkedIn and subscribe to our newsletter.