LIMA Desktop¶
LIMA Desktop is a Windows-based AI screen reading assistant designed for blind and low-vision users. Unlike traditional screen readers that rely on fixed, pre-programmed output, LIMA uses advanced AI to understand and describe what's on your screen in natural, conversational language.
Note
LIMA Desktop was previously called LIMA Screen Reader.
Version
This documentation covers LIMA 2.0.0.0, released on 20 October 2026.
You control LIMA with your voice or keyboard — no mouse required.
How LIMA Fills Gaps Left by Traditional Screen Readers¶
Tools like NVDA and JAWS are excellent at reading structured content — menus, form fields, and document text. But they struggle with:
- Inaccessible visuals — charts, graphs, images, and buttons and controls that apps draw themselves and never label
- Dynamic web content — modern websites that update themselves as you use them, without reloading the page
- Natural language understanding — you must learn exact commands rather than asking naturally
- Context-aware descriptions — traditional readers read text literally; they don't interpret meaning
LIMA addresses these gaps directly:
| Capability | NVDA / JAWS | LIMA |
|---|---|---|
| Structured document reading | Yes | Yes |
| Natural voice commands | No | Yes |
| AI image & chart description | No | Yes |
| Dynamic web content | Limited | Yes |
| Multiple input methods | Keyboard only | Voice + keyboard |
| Supported languages | Many | English, Vietnamese, Tagalog |
Next Steps¶
Using these docs with an AI assistant¶
All of the LIMA Desktop documentation is also available as a single text file: LIMA Desktop documentation for AI assistants. Give this link to an AI assistant and it can read every page at once, rather than one page at a time.