Elephas review

Elephas is a document question-and-answer app for Mac and Windows. It indexes files on your computer so you can ask questions across them instead of opening each one and searching manually. Supported formats include PDFs, Word documents, spreadsheets, slide decks and scanned pages.

Answers cite the source file and show the quoted passage, along with its page, slide or sheet. Elephas can search in English, German, Spanish, French, Japanese and Chinese, including across languages. You can use Elephas AI, connect your own API key or run a model on your computer.

For privacy, Elephas can mask names, email addresses and phone numbers before questions and passages leave your computer. Teams can share a knowledge base through Team Brain, with role-based access and end-to-end encrypted sync. That does not mean every AI request stays on-device: where the content goes depends on the AI option and settings you choose.

The clearest value is being able to ask questions across a collection of work documents and check the answer against a citation. Elephas shows the quoted source passage and its location, giving you something to verify rather than asking you to take a summary on trust. Support for scanned pages also makes it useful beyond clean, text-based office files.

The choice of AI setup is another practical plus. You can use Elephas AI, bring an API key or run an on-computer model. Personal detail masking adds a targeted privacy control: the product says it hides names, email addresses and phone numbers before questions and passages leave your computer. It is not a blanket promise that all processing stays local.

For teams, Team Brain adds shared knowledge bases, role permissions and end-to-end encrypted sync. The pricing information says there is no minimum team size, and AI usage is pooled across seats. That may suit small teams that want shared document answers without imposing one AI setup on every user. Six-language search, including cross-language search, is useful for multilingual document collections.

The biggest buying uncertainty is AI usage. Personal includes a monthly amount of Elephas AI, and Professional includes twice as much, but the exact limits are not published. Usage resets monthly and does not roll over. When included usage runs out, it cuts off or you switch to your own API key or an on-computer model; the pricing information lists no overage fees.

The pricing page lists Personal at $19 monthly or $15 per month billed annually ($180 a year), limited to one computer. Professional is $29 monthly or $24 per month billed annually ($288 a year), for up to three computers. Team is listed at $29 per person per month, but its annual price is not published. There is no free plan; the listed trial is seven days.

User feedback on Reddit is mixed. One commenter called Elephas confusing to figure out, while another said it felt undercooked. A separate thread reported crashes and intermittent sync, though those reports are not proof of current product behavior. We would treat them as a reason to use the trial to check the workflow on your own files and devices.

Elephas has no free plan; the pricing information lists a 7-day trial. Personal costs $19 monthly or $15 per month billed annually ($180 upfront for the year). It covers one computer, with a monthly AI allowance whose exact limit is not published. Professional is $29 monthly or $24 per month billed annually ($288 a year): $10 more month to month, or $9 more per month on annual billing. It covers three computers and twice Personal’s included AI usage, but the exact allowance is also not published.

Team is listed at $29 per person per month, with no minimum team size, but an annual price is not published. The owner and pending invitees count as seats. Team AI usage is pooled. No paid add-ons or overage fees are published. When included usage runs out, Elephas says usage cuts off or switches to your own API key or an on-computer model. Personal and Professional usage resets monthly and does not roll over.

Moving to Elephas means indexing your existing files; the product description lists PDFs, Word documents, spreadsheets, slide decks, and scanned pages. It does not specify how long indexing takes or whether you can import an existing knowledge base, so check those details before planning a team rollout. You can use Elephas AI, your own API key, or an on-computer model, which gives you alternatives if you stop using its included AI allowance.

Leaving is less clear. The published materials do not publish a data-export process or explain how to move Team Brain content, permissions, or citations elsewhere. Confirm what you can export before building a shared knowledge base around it. We also don’t have attributable Reddit posts or comments in the published material for this review, so we can’t responsibly say what users there report as reasons for leaving. Ask about export and retention directly rather than assuming local indexing makes the whole setup portable.