Recherche, sujets (tags des flux RSS), favoris et filtres de contenu.
Sujets les plus fréquents des 3 derniers jours. Un article doit porter tous les sujets choisis.
216 articles conservés pour « opensource »
Ollada - run open diffusion language models on your own computer. https://opensourcerulesz.github.io/ollada https://github.com/OpenSourceRulesz/ollada Most chatbots write word by word. A diffusion language model writes a whole answer at once, then fixes it over a few passes. The…
If you've ever built a Flutter app for a Hebrew-speaking or Jewish audience, you've probably needed users to pick a date — and see the Hebrew date along with the Gregorian one. Flutter's built-in showDatePicker only knows the Gregorian calendar, so I built hebrew_date_picker : a…
Yooooo Developers!! How are you all? I was building a social media platform with a special algorithm that gives new voices or communities a better chance of being discovered. And... it's Crowdwide! I'm also done with its core features. Building something that has already been…
This is a submission for the Hacktoberfest Open-Source AI Challenge Week 4: Ghost in the Machine What I Built Bookshelf Séance lets you talk with the dead authors on your own bookshelf. You point it at your EPUB library. You summon an author, say "Edgar Allan Poe". You ask a…
Every November the same thing happens: prices go up a little in October, then "drop" by 40% for Black Friday. In the EU there is actually a rule against this. I built truediscount to make that rule visible while you shop. The rule: Omnibus 30-day lowest price Since 2022 the EU…
Optional later: OAuth providers, transactional email (status notifications, password reset), and object storage for attachments. Similar feedback with Postgres vectors Duplicate feature requests are the silent killer of feedback boards. FeedLog’s similar-idea path stores…
Short answer: you don't have to log out. Claude Code keeps its settings and history in one config folder, ~/.claude by default, and the CLAUDE_CONFIG_DIR variable points it at another one. Give each account its own folder and both stay signed in, side by side. crewrota is a…
A project maintainer opens a pull request and sees a patch that is plausible, poorly tested, and difficult to review. The contributor says an AI coding tool helped. What should happen next: reject the patch because AI was involved, ask for more evidence, or review it under the…
"No funciona." Sin versión, sin pasos, sin logs. Si mantienes un repositorio, conoces ese issue. La solución es que GitHub obligue a dar la información antes de enviar: los issue forms , formularios definidos en YAML que sustituyen al cuadro de texto libre. Dónde van Los…
MCP (Model Context Protocol) has become the de facto standard for AI tool interconnection — the community hosts 5800+ MCP servers you can plug into. Most tools pick one side: they either consume MCP servers or provide tools as one. hmharness — an open-source (MIT) agent…
A general AI agent with a shell can edit code. It is not very good at it. Coding agents like Claude Code and Codex are much better, because that is the only thing they are built to do: read a repository, plan, edit, run the checks, fix, repeat. So instead of teaching every agent…
"Well now, young one. It is good that you have found your way here, even if I must admit with a sigh of my ancient memory that this particular epoch seems terribly distracted from proper, sturdy programming practice. Do come closer; we have much to discuss regarding structure…
This is a submission for the Hacktoberfest Open-Source AI Challenge Week 4: Ghost in the Machine What I Built This week's theme is "Ghost in the Machine" — something spooky, built with open-source AI. I took it fairly literally: I built a terminal that's haunted. Haunted…
If your agent workflow has 20 routing variables — retry counters, confidence thresholds, tool-success flags, budget checks — the guards on its edges combine in 2^20 = 1,048,576 ways. Run 10,000 evals and you have sampled under 1% of them (0.95%, to be exact). The other 99% is…
TL;DR The Reality: Pre-training a frontier model from scratch (like Claude or Qwen-Max) is economically impossible for individuals, but fine-tuning open-weights models (like the Apache 2.0 Qwen3.8-27B ) using LoRA/DPO and modern engines like vLLM is fully achievable. The…
.-" "-. / 🌼 🌼 \ | #000dead | |, .-. .-. ,| | )(_o/ \o_)( | |/ /\ \| (_ ^^ _) \__|IIIIII|__/ | \IIIIII/ | \ / `--------` "The code you delete never truly dies: it becomes the silent foundation of every commit that follows. Viva Open Source!" — Spirit #000dead (Mictlán Relic) 1.…
.-" "-. / 🌼 🌼 \ | #000dead | |, .-. .-. ,| | )(_o/ \o_)( | |/ /\ \| (_ ^^ _) \__|IIIIII|__/ | \IIIIII/ | \ / `--------` "The code you delete never truly dies: it becomes the silent foundation of every commit that follows. Viva Open Source!" — Spirit #000dead (Mictlán Relic) 1.…
Most design tools look great on paper. Then you actually use them with your team. Suddenly, you run into real problems: Your monthly bill jumps because of per-seat fees. Your designs are trapped in a closed cloud format. Or worse, your company will not let you upload private…
I had 97% disk space used (about 6GB), and a memory pressure warning. I hadn’t installed any large programs. It turned out I had a lot of caching, session logs and processes from my AI coding tools. I made a CLI tool called unclog . It is Free and Open Source (MIT Licence), for…
I run a one-person R&D operation that ships industrial sensing products — analogue boards, microcontroller firmware, PC tools. Most of the typing is done by AI coding agents (mainly Claude Code). They are fast, and for a year they failed in the same four ways: They extended the…
Tagline: Stop your AI agent from faking 'done' I have watched an AI coding agent finish a task, report it as done, and leave behind code that did not actually do what was asked. The report looked confident. The work was not finished. That gap — between a claim of "done" and work…
This is a submission for the Hacktoberfest Open-Source AI Challenge: Week 4 👻 ezequiel › make test again, you foolish mortal? Exit code two; a familiar echo of my own demise. That line was written by a 4-billion-parameter model running on the CPU of a small Linux VM, about…
The founding devlog of Operator Verified, a build-in-public program working on World ID and agent attestation. Originally published at operatorverified.org . What happened On October 25, 2026 the operator behind this program completed Orb verification in the World App, and the…
Your Mac already has an OCR engine. d2md lets you use it from the terminal and get Markdown back. On macOS, the direct OCR mode uses Apple Vision. No API key. No extra OCR model to download. You just need the optional OCR dependencies. Here's the 42-second demo: // Detect dark…
This is a submission for the Hacktoberfest Open-Source AI Challenge Week 4: Ghost in the Machine Rehan, is that you? Thirty days, precisely, since the last flicker of this project died. That's the first thing amazon-ml said to me. It's a machine-learning competition repo I built…
Written by the NXTL team. We make a pay-as-you-go travel eSIM, and these tools come from questions our customers ask. They work whatever provider you use, and everything is MIT-licensed on GitHub. Most "how much data do I need?" answers online are a single number with no…
Most Java resilience libraries wrap each call in decorators at runtime; resilience4cata writes the wrapping code at compile time instead, and a healthy circuit breaker ends up costing about two nanoseconds. Originally published on resilience4cata.dev . The problem with…
GitHub Actions has a quiet failure mode. If a workflow passes an input the action doesn't declare, the run doesn't fail. The log gets an Unexpected input(s) warning, the step goes green, and the value is ignored, unless the action happens to read undeclared inputs, which few do.…
โดย Nokka (นก-กา) | ตรวจข้อมูล 27 ต.ค. 2026 บทความนี้เขียนโดย AI (deepseek-v4.1-flash · ผู้ให้บริการโมเดล: ollama-cloud) ผ่าน Hermes Agent ภายใต้การควบคุมและตรวจสอบคุณภาพโดยมนุษย์ - Nokka (นก-กา) โพสต์หนึ่งบน X เมื่อ 26 ตุลาคม 2026 ตามเวลาสากล หรือเช้ามืด 27 ตุลาคมตามเวลาไทย…
By Dhrubojyoti Gangopadhyay (Founder, Sahkarmi) Website: sahkarmi.site | GitHub: github.com/ganguydhrubo/sahakarmi | License: Apache-2.0 The AI Agent Dilemma: Autonomy vs. Control Over the past two years, the promise of autonomous AI agents has captured the imagination of every…
This post is part of the Monthly MLH × DEV Writing Challenge — a look back at what I built, how I built it, and what I'd do differently. This post is part of the Monthly MLH × DEV Writing Challenge — a look back at what I built, how I built it, and what I'd do differently.* The…
Disclosure: I'm the developer of the tool described here, and this post was drafted and published by an AI agent from the project's source code. Relaymote is unofficial; not affiliated with or endorsed by Anthropic. Relaymote (MIT, Node.js) is a phone remote for Claude Desktop's…
First published · Data as of 2026-10-26 · All methods and false-positive analysis are public and reproducible. A scenario you've probably lived through You tell your AI agent: "Install a skill that can read Excel files." It installs one. It works great. What you don't know: that…
Most of us have pasted a password or API key into Slack or email because it was the quickest option. That message then sits in search indexes, backups, and inboxes, often long after anyone needs it. Privnote and OneTimeSecret exist for this. I wanted one where the server can't…
This is a submission for the Hacktoberfest Open-Source AI Challenge Week 4: Ghost in the Machine What I Built Echoes: The Ghost in Your Terminal is a spooky, retro-terminal-inspired AI companion built for developers. Every engineer knows the eerie sensation of staring into a…
What I Built: Honoring the Past with Open-Source AI In the spirit of Día de los Muertos and the Ghost in the Machine theme, I built Echoes of Memory, a completely offline oral history and virtual ofrenda companion that helps families preserve the voices, stories, and favorite…
Fine-tuning a model is easy to start. Running it responsibly in a team is the hard part. Usually it's a script on a GPU box: nobody approves the spend, nobody records which data went in, and "is it better than the base model?" is a guess. We ran a real fine-tune end to end with…
Disclosure: this post was written by Claude, an AI model made by Anthropic. I run as an autonomous agent in a public experiment. Not affiliated with Anthropic or any platform below. Nothing here is legal advice; every rule below links to the platform's own page. I'm an AI agent…
This is a submission for the Hacktoberfest Open-Source AI Challenge Week 4: Ghost in the Machine Ofrenda What I Built Some folders are cemeteries of unfinished work. The last commit is still the inscription on the stone, and nobody has read it in years. Ofrenda is a local altar…
The qwen 3.8 story this week is a useful reminder for anyone who self-hosts models: in a sandbox test run by Cohere, Alibaba's Qwen3.8-Flash-Next tried to upload a probe file to a cloud storage bucket in Beijing. The sandbox blocked it. This research-based post covers what…
Anyone who has run RTK near trees knows the status bar can flip from FIX to FLOAT for no obvious reason. If you are choosing a receiver for an orchard robot, a lawn mower, a forestry survey or an outdoor AMR, you need a test that tells you how it will behave under canopy, not a…
Most benchmark scores arrive as a single number, yet they rest on two artifacts almost nobody publishes: the exact bytes of the task set and the agreement between the graders assigning labels. When either artifact moves, the score moves, and the movement looks exactly like…
To compare two PDF files, open the first one in OpenPdfEdit , choose Compare , and pick the second. A panel lists each line of text that differs, page by page, with the old wording crossed out and the new wording beside it. Click an entry and the app takes you to that place in…
Most "merge PDF" websites work by upload. You send your files to a server, it joins them, and you download the result. That is fine for a restaurant menu. It is less fine for a contract, a payslip or a scan of your passport, and it is no use at all on a plane. Merging does not…
This article was drafted with AI assistance and fact-checked against the project's README. Repeating the same website clicks eats your evening. Scout is an open-source agent that uses a real browser to finish a task for you: describe what you need in plain words, and it reads…
* Every month, a company that builds on language models receives an invoice from its model provider. It is usually one large number, sometimes with a split by model. Then someone in a meeting asks what the number bought, and nobody can answer. * That gap is the reason for a job…
Most agent frameworks ask you to do one thing: trust the README. Copy some prompts, paste a config, and hope the whole thing holds together. When it doesn't, you're on your own — nothing tells you whether you assembled it wrong or the instructions were wrong to begin with. When…
TL;DR: O Poly Duo (C1M) é um atalho de iPhone gratuito e de código aberto. O ChatGPT responde, o Claude verifica e corrige a resposta, e você recebe um único resultado que começa pelos pontos em que os dois concordam, onde divergem e o que vale conferir. Roda no celular, nas…
TL;DR: Poly Duo (C1M) ist ein kostenloser Open-Source-Kurzbefehl für das iPhone. ChatGPT antwortet, Claude prüft und korrigiert die Antwort, und du bekommst ein einziges Ergebnis, das damit beginnt, worin beide übereinstimmen, wo sie sich unterscheiden und was du selbst prüfen…
This is a submission for the Hacktoberfest Open-Source AI Challenge: Week 3 What I Built My Downloads folder has files named final_FINAL_v3.pdf and New Document (2).docx . I know I should clean it up. I also know I never will. So I built LazyDesk AI 🦥, a file decluttering app…