Productivity●●Solid
Local CPU OCR for images, PDFs, webpages
CPU-only OCR with clipboard in/out beats Tesseract for modern screenshots.
Ship ItSolve My Problem
mrkn1
301mo ago
Back to browse
GitHub Repository
Snap any image, screenshot, or webpage into plaintext. No GPU. No cloud. One command.
174 starsPython
Local-first fast CPU image to text for screenshots, PDFs, webpages
by mrkn1·Jun 5, 2026·19 points·17 comments
AI Analysis
●●SolidCozySolve My Problem
CPU-only OCR with clipboard round-trip when cloud APIs dominate the space.
Strengths
- •Quantized 0.9B ONNX model runs on plain CPU without GPU or M-series requirements
- •Clipboard-in-clipboard-out workflow makes screenshot-to-text a two-step process
- •Single Python module with self-installing dependencies and portable folder deployment
Weaknesses
- •890 MB initial download may deter users on limited bandwidth
- •Webpage URL handling OCRs single image rather than full page text extraction
Category
Target Audience
Developers and power users needing offline text extraction
Similar To
Tesseract · Google Cloud Vision · ABBYY FineReader
Similar Projects
Developer Tools●●Solid
CPU-only fast OCR for screenshots, images, PDFs, webpages
CPU-only VLM OCR beats Tesseract accuracy without sending data to the cloud.
Solve My ProblemCozy
mrkn1
981mo ago
AI/ML●●Solid
CPU-only OCR for screenshots, images, and webpages
CPU-only VLM OCR beats Tesseract on layout without needing CUDA or cloud APIs.
Solve My ProblemCozy
mrkn1
491mo ago
Developer Tools●Mid
ocrbase – PDF/IMG –>.MD/JSON Model-Agnostic OCR API
Yet another OCR API wrapper when JinaAI and Firecrawl already exist.
Ship It
adammajcher
103mo ago
AI/ML●●●Banger
fftext - quick summary, fact check, ELI5 on CPU
Fact-check command extracts claims and cites sources—rare for local AI.
Dark HorseBig Brain
mrkn1
101mo ago
AI/ML●●Solid
Free summary, fact check, ELI5 of any text on CPU
Fact-check mode extracts claims, web-searches each, and labels them with citations — rare for local LLMs.
CozyNiche Gem
mrkn1
201mo ago