Local model inference
Choose a compatible model, review its size and device requirements, then download it deliberately. Normal chat generation runs on your device rather than a CuriousLM inference server.
Private local AI
Useful AI, free and private. AI chat runs 100% on your device
Your prompts, conversations and files stay on your device
Internet is still needed for first load, model downloads, updates, and optional confirmed network features.
The short answer
What does local AI mean?
Local AI performs inference on the device you are using. The model weights and runtime are downloaded first; after that, a prompt can be processed without sending it to a hosted model API. This differs from a private cloud chatbot, which may protect data in transit or limit retention but still computes the answer on remote servers.
Built for everyday work
More than a local model playground.
CuriousLM combines on-device generation with the working context people expect from a modern assistant.
Files and citations
Bring supported documents and images into a project. CuriousLM extracts and retrieves relevant passages locally and keeps citations connected to their source.
Projects and memory
Keep related chats, local files, project instructions, model choice, and optional memories together instead of rebuilding context in every conversation.
No account or advertising
Open the app and use it without creating a CuriousLM account. There are no ads. Cookieless aggregate usage statistics never include chat or file content and can be turned off in Settings.
A clear network boundary
Local content by default. Online activity is clearly disclosed.
Chats, projects, memories, imported files, indexes, and downloaded models are stored on your device. The PWA protects private records and files with its local vault; Android uses encrypted native storage protected by Android Keystore and device authentication.
- Normal local chat does not call a CuriousLM inference API.
- Model downloads are deliberate; required runtime components can arrive with the app, while optional runtime packs are downloaded separately.
- Optional web search shows the query before sending it to Tavily.
- Response reports show the exact excerpts before submission.
- Cookieless aggregate analytics never includes chat or file content and can be turned off in Settings.
Choose the right architecture
Local AI and cloud AI solve different problems.
| Question | On-device AI | Hosted cloud AI |
|---|---|---|
| Where is inference performed? | Your phone or browser | A provider's servers |
| Can normal chat work offline? | Yes, after setup | Usually no |
| What limits model size? | Device memory and storage | Provider infrastructure and plan |
| How current is built-in knowledge? | Limited to the downloaded model | May include newer models and connected tools |
| Who controls the model files? | The user's device | The service provider |
Privacy depends on the complete implementation, not the word “local” alone. Read our detailed local AI vs cloud AI guide.
Models you can inspect
See the cost before you download.
Local models vary significantly in storage, memory use, modality, licence, and speed. CuriousLM keeps the catalogue visible, explains device requirements, verifies downloaded artifacts, and makes activation explicit. Candidate entries remain labelled until their platform qualification is complete.
How to choose a model for Android
Common questions
Private local AI, explained plainly.
Does CuriousLM work without the internet?
After the application shell and a compatible model have been downloaded, normal local chat can work offline. Initial installation, model downloads, updates, and optional web search still require a network connection.
Are prompts sent to a CuriousLM server?
Normal local-model prompts and responses are processed on the device. Optional Tavily web searches and response reports send only the information shown in a confirmation step after you approve the action.
Which files can local AI work with?
CuriousLM supports common text and office formats including PDF, DOCX, PPTX, XLSX, CSV, HTML, Markdown, and plain text, plus supported images. Extraction quality varies with the source file, OCR quality, and selected model.
Will every phone run every local model?
No. Model compatibility and speed depend on memory, storage, graphics support, operating system, and the model itself. CuriousLM shows model size and device requirements before download and does not hide unavailable catalogue entries.
No account required
Try private local AI in your browser.
Open CuriousLM, choose a compatible model, and keep control of where normal chat runs.
Open CuriousLM