Private AI Chat
elmtree chat is an AI chat interface for people who want capable models without organizing their work around data collection.
We build elmtree around a simple rule: your content should only be processed for the function you are using right now. Not for training, not for advertising, not for profiling, and not for product analytics.
What elmtree is for
- AI chat with current hosted language models.
- Document work with preview and pseudonymization.
- Web search when current information or sources are needed.
- Clear monthly plans with fair-use allowances.
Privacy as a product feature
elmtree does not store chats as readable server history. We do not analyze your chat messages. Requests to LLM providers are routed so they cannot be associated with your elmtree account.
When you work with documents, pseudonymization helps replace detected personal data with tokens before text is sent to a model. The preview shows which version will be sent.
Documents and large files
Short and simple documents can be processed locally in the browser. For larger or technically difficult documents, optional server processing can help. This processing serves the concrete task: extracting text, running OCR, preparing document sections, or enabling RAG.