标签：#ocr

Introduction I'm the author of TrulyFreeOCR, an open-source OCR pipeline that turns scanned PDFs into searchable, highly-compressed PDFs. Everything is Apache 2.0 / MIT / BSD — no GPL, no AGPL, no proprietary model weights. Why I built it: I needed an OCR pipeline for a document processing system where: Every dependency had to be business-friendly (no GPL/AGPL) Deployment required zero admin rights (no sudo, no brew, no apt-get) MRC compression was needed to hit 5-10x file size reduction vs JPEG-only Everything had to run offline on CPU — no cloud APIs, no GPU I surveyed 20+ existing tools (full comparison in the repo's docs) and none fit all requirements. OCRmyPDF is closest but needs Python + Ghostscript + Tesseract as system deps, and MPL-2.0 requires publishing modifications. The VLM models (DeepSeek-OCR, GLM-OCR, etc.) produce better text extraction but need GPUs and don't output PDFs at all. What it does: Input: any PDF (scanned, born-digital, or mixed) Output: searchable PDF with invisible text layer + MRC compression (JBIG2/CCITT foreground + JPEG background) Single fat JAR — one file to copy, one command to run Bootstrap script downloads everything (JDK, Gradle, Tesseract, Leptonica, jbig2enc) into project subdirs Fully offline, CPU-only PDF/A-2b output available 7 bundled language models, 100+ more downloadable Concurrent OCR (configurable thread pool) Try it in 3 commands: $ git clone https://github.com/msmarkgu/TrulyFreeOCR.git $ cd TrulyFreeOCR $ ./bootstrap.sh ./run.sh tests/simple-text.pdf -o output.pdf Limitations (being upfront): Tesseract-based accuracy — good for clean scans, not SOTA for noisy/photographed docs No table/formula extraction yet No handwriting recognition CPU-only is slower than GPU backends for high volume Would love feedback — especially from anyone who's tried to deploy OCR in an enterprise environment. https://github.com/msmarkgu/TrulyFreeOCR

How to Create an AI Agent: A Production Walkthrough The first agent I shipped to production failed at 3am on a Sunday. It looped on a tool call, burned through $40 in tokens before my budget alarm fired, and left a half-written draft in the database with no way to resume. That night taught me more about agent design than any framework tutorial. Since then I have built a pattern I trust enough to leave running unattended for weeks at BizFlowAI, where agents research, write, optimize and publish content without me touching them. This is that pattern, stripped down to what actually matters. Start with the job spec, not the framework Before you pick LangGraph, CrewAI, or roll your own, write the agent's job spec like you would for a junior engineer. One paragraph. What it owns, what it must never do, what "done" looks like, and which signals tell you it failed. Here is the spec for one of my production agents: The Topic Researcher owns generating a ranked list of 20 content topics per site per week. It reads from keyword_pool and search_console_perf , writes to topic_queue . It must never publish, never call paid APIs more than 8 times per run, and must finish in under 6 minutes. Done = 20 topics with score >= 0.6 and zero duplicates against the last 90 days. Failure signal = empty queue after a run, or any topic flagged by the dedupe check. If you cannot write this paragraph, do not build the agent. You will end up with a "do everything" prompt that hallucinates its way through ambiguous tasks. The job spec becomes your evaluation rubric later, so write it carefully. Rule of thumb I use : if the spec needs more than 5 tools or more than 3 decision branches, it is two agents, not one. Design the tools before you write the prompt Most agent failures I have debugged were not prompt failures. They were tool failures. The model called a tool with wrong arguments, the tool returned a 4MB JSON blob, or two tools had overlapping responsibilities and the model picked the wrong

AI Dev Weekly is a Thursday series where I cover the week's most important AI developer news, with my take as someone who actually uses these tools daily. OCR had a week. Mistral dropped OCR 4 with bounding boxes. Baidu open-sourced a model that beats DeepSeek-OCR. Claude got a permanent home inside Slack. And the Fable 5 ban fallout keeps getting uglier: Alibaba was apparently stealing Claude's capabilities, and even the NSA lost access to Mythos. Meanwhile, GPT-5.6 is delayed to mid-July. Let's go. 1. Mistral OCR 4: document AI gets serious Mistral launched OCR 4 this week. It's not just another OCR model. It's a full document understanding system with paragraph-level bounding boxes, confidence scores, and support for 170 languages. The specs: $4 per 1,000 pages (standard), $2 per 1,000 pages (batch) Paragraph-level bounding boxes with coordinates 72% win rate in blind tests against competitors Available on la Plateforme, Microsoft Foundry, and self-hosted for enterprise Top score on OlmOCRBench Why this matters for developers: Bounding boxes change everything. Previous OCR models gave you text. Mistral gives you text + where it is on the page. That unlocks document search, compliance systems, and any workflow where page structure matters. My take: At $4/1000 pages, this is competitive with Google Document AI ($5) and significantly cheaper than building your own pipeline. For enterprise document processing, this is probably the best option right now. For budget-conscious developers, Baidu's free alternative (see below) is worth considering. Full comparison in our Mistral vs DeepSeek vs Baidu breakdown. 2. Baidu open-sources Unlimited-OCR While Mistral went commercial, Baidu went open. Unlimited-OCR is a 3B-parameter MIT-licensed model that processes multi-page PDFs in a single inference pass. Key features: Built on DeepSeek-OCR architecture (SAM+CLIP + DeepSeek-V2 MoE decoder) Reference Sliding Window Attention for memory efficiency on long documents Tables to HTM

Paperless-ngx is an open-source document management system that converts scans and PDFs into a fully searchable archive using Tesseract OCR, with tags, custom fields, and automated processing rules. This guide deploys Paperless-ngx using Docker Compose with PostgreSQL, Redis, and Traefik handling automatic HTTPS, then uploads a document and verifies OCR extraction. By the end, you'll have Paperless-ngx serving an OCR-indexed document archive securely at your domain. Set Up the Directory Structure 1. Create the project directories: $ mkdir -p ~/paperless-ngx/ { data,media,export,consume,pgdata } $ cd ~/paperless-ngx 2. Create the environment file: $ nano .env DOMAIN = paperless.example.com LETSENCRYPT_EMAIL = admin@example.com PAPERLESS_SECRET_KEY = CHANGE_TO_RANDOM_STRING PAPERLESS_URL = https://paperless.example.com PAPERLESS_TIME_ZONE = UTC PAPERLESS_OCR_LANGUAGE = eng PAPERLESS_DBPASS = STRONG_PASSWORD_HERE POSTGRES_DB = paperless POSTGRES_USER = paperless POSTGRES_PASSWORD = STRONG_PASSWORD_HERE Use the same value for PAPERLESS_DBPASS and POSTGRES_PASSWORD . PAPERLESS_SECRET_KEY should be a 32+ character random string. Deploy with Docker Compose 1. Create the Compose manifest: $ nano docker-compose.yaml services : traefik : image : traefik:v3.6 container_name : traefik command : - " --providers.docker=true" - " --providers.docker.exposedbydefault=false" - " --entrypoints.web.address=:80" - " --entrypoints.websecure.address=:443" - " --entrypoints.web.http.redirections.entrypoint.to=websecure" - " --entrypoints.web.http.redirections.entrypoint.scheme=https" - " --certificatesresolvers.letsencrypt.acme.httpchallenge=true" - " --certificatesresolvers.letsencrypt.acme.httpchallenge.entrypoint=web" - " --certificatesresolvers.letsencrypt.acme.email=${LETSENCRYPT_EMAIL}" - " --certificatesresolvers.letsencrypt.acme.storage=/letsencrypt/acme.json" ports : - " 80:80" - " 443:443" volumes : - " ./letsencrypt:/letsencrypt" - " /var/run/docker.sock:/var/run/docker.sock:ro"

On 14 April, the Trump administration quietly acknowledged the widespread use of AI to automate government processes. The office of management and budget (OMB) disclosed a staggering 3,611 active or planned use cases for AI across the federal government. The list has ballooned by 70% from the one published in the final year of the Biden administration, and includes many disturbing-seeming plans to hand over sensitive governmental functions to AI. Scanning this list, many readers may find many causes for alarm. It represents a transfer of decision processes from human to machine on a massive scale over matters of individual freedom, public health and well-being, nuclear reactor safety and more...

Let no one accuse Bernie Sanders of ducking the big questions. Writing in the New York Times last week, the senator asked : “Will the future of humanity be determined by a handful of billionaires who have promoted and developed AI, with virtually no democratic input, who stand to become even richer and more powerful than they are today?” We agree entirely that this is one of the most potent questions facing global democracy today. Our book, Rewiring Democracy , surveys the emerging uses for and impacts of AI in democracy around the world and reaches the same conclusion: that the most urgent risk posed by AI is the ...

Younger Americans have soured on the second Donald Trump presidency , but they are not protesting it. Despite an unpopular Iran war and an even more unpopular Trump administration , college campus protests nationwide have gone silent . And at many schools, student activism is virtually nonexistent . This silence comes in the wake of a relentless Trump administration war on campus speech that has involved lawsuits , arrests , deportations and expulsions . Reports cite a range of complicated factors for the restraint, from apathy to technology-induced incapacity. But as ...