All AI Pulse updates tagged "Open Source".
66 signalsUpdated the DeepSeek Harness plugin for Baoyu Design Skill, now with a unified entry point to view all generated Slides, Design, Dashboard, and Motion results, and also convenient for initiating corresponding sessions. Update address: https:// github.com/JimLi
Google's open-source SDK lets TypeScript and JavaScript developers call Gemini models to generate text, images, and other content.
With AI, writing code has gotten way faster, but verification is still slow, especially waiting on GitHub Actions CI. I spent this afternoon giving the CI for Mole, Kaku and Pake a full overhaul, and the results were great, so I want to share what I learned. J
I replicated this with Claude Code! It took a live webcam 7 km south of Southampton, UK and predicted its location purely from doppler effect in audio spectrogram + open-source aircraft location data. Lo and behold, the predicted location (heatmap) peaks at ac
My next mission is to direct tokens at useful and meaningful problems in life 💪 Need more tokens!!! Anyone know how to get frontier level tokens for open-source projects? 😅
A project detects spoken wake words from live mic audio and triggers apps entirely on device, targeting microcontrollers, Raspberry Pi, phones, browsers, and desktops.
Open-source AI application runtime for reliable, observable, and governable production AI systems, helping developers deploy AI applications to production with confidence.
Public OpenCAS runtime snapshot and release docs featuring a local-state agent runtime, dashboard controls, memory, scheduling, channels, and provider-routed model access.
Modular AI runtime platform for operational intelligence with RAG, orchestration, tooling, governance, and observability. Built on FastAPI, PostgreSQL + pgvector, Ollama, and enterprise DevOps integrations like Jenkins and Nexus.
yetone thanked the community as Magpie's third-party plugins hit 115 total, 93 from vendors, showing steady ecosystem growth for the open-source project.
Grep has open sourced big parts of its harness, which Garry Tan calls a big deal for the developer tooling ecosystem.
Anthropic donated Agent Skills, its open standard giving AI agents specialized knowledge and workflows, to the Linux Foundation's Agentic AI Foundation. Since December 2025, over 70 platforms including ChatGPT, Cursor, and VS Code have integrated it, with 1 mi
The poster argues DeepSeek open sourcing helps smaller firms afford infrastructure talent once limited to big tech, widening access to Infra engineering skills and jobs.
openagents-org released launcher-v1.1.2, with connector 0.2.196 adding saved browser tab reuse, browser sessions and live thread previews, plus fixes for sign-in, device pairing, and onboarding.
Google open-sourced Harness², a way to improve AI agents without touching the model. The agent rewrites its own harness while the model weights stay frozen.
A developer shared experience using MiniMax H3 for Douyin videos and open-sourced reelbench-skills/video-scrub to remove metadata. Their first three videos without scrubbing got noticeably fewer views.
Google's open-source AI agent that brings the power of Gemini directly into your terminal for command-line workflows.
A viral skill and proxy for coding agents that cuts 65% of tokens by talking to models in a minimalist caveman style.
LangChain positions itself as the agent engineering platform, offering a full toolchain for building, debugging, and deploying AI applications.
An open collection of system prompts, internal tools, and AI models for Cursor, Claude Code, Devin, Manus, Windsurf, and many other AI tools.
ComfyUI is a powerful, modular diffusion model GUI, API, and backend with a graph/nodes interface, billed as the fastest local inference engine.
Build agentic workflows and RAG pipelines in one collaborative workspace. Deploy on cloud, VPC, or self-hosted, moving teams from prototype to production.
AutoGPT aims to make AI accessible for everyone to use and build on, providing the tools so users can focus on what matters.
Formerly Awesome ChatGPT Prompts. A community platform to share, discover, and collect prompts. Free, open source, and self-hostable for full privacy.
User-friendly AI Interface (Supports Ollama, OpenAI API, ...)
The author wants the report as an OSS self-serve skill, using GitHub distribution to reach AI engineers and create a growth loop.
Send this video to Grok and let it make a game out of it. It was implemented with a total of just 1500 lines of code. Try it out: https:// wy51ai.github.io/roadster-game/
openEuler says AI is scaling and the OS is part of the story. The project will appear at Open Source Summit + Embedded Linux Conference Europe 2026, where AI infra talks are in full swing.
A developer discovered REA, an open-source project claiming to reverse-engineer any software. Spot a useful feature in someone's app and have an agent replicate it, sparking debate about whether indie devs still need to build products themselves.
OpenAI publishes new results on open mathematics problems from an internal frontier model, sharing Lean proof formalizations and research details on GitHub.
SuperPlane is an open source tool that has AI agents turn routine backlog tasks into reviewed pull requests, keeping judgment calls with your team.
Developer hylarucoder open-sources an AI skill, describing it as one of the few non-technical skills on the market written specifically for technical personnel.
Arav Srinivas tweeted "Never bet against open weight models," expressing strong confidence in the open weight model ecosystem.
A new open-source tool tracks objects across video frames using any detector like YOLO or DETR, keeps IDs stable during camera movement, and is free even for closed-source products.
$399, open-source, and it can roller skate. Hugging Face releases the small robot Microduck: it can walk, pick up objects, get back up after falling, and even learn new actions through reinforcement learning. Hands-on exploration of embodied AI and world model
512GB of DDR5. That is the line nobody screenshotted from the FreeToken paper. Berkeley and MIT open-sourced it yesterday and the headline numbers are real: 753B GLM-5.2 on one workstation GPU at 14.9 tok/s, 35B on an 8GB laptop at 39.3. I installed it on my 3
THIS IS ABSOLUTELY INSANE a developer just released a free tool that builds a vertical video with a talking presenter. out of one script and one photo. it's called lanshu-create-ai-presenter-video. it landed on GitHub two days ago. 267 stars. MIT. here's how i
Origin, our code hosting platform, is now live. It's fast, easy to use, and deeply integrated with Cursor. Get started by syncing your repos from GitHub.
There are already open-source projects for removing text watermarks https:// github.com/guillaumemeyer /watermarks-remover …
开放权重会毁掉AI创新吗？ 1/ AI开源与闭源之争，可能从一开始就问错了。 Christian Catalini梳理大量经济学研究后提出：开放权重未必减少AI投资，只会改变【钱投向哪里、谁负责创新、最后谁赚走利润】。 这个视角挺值得看。
Meta's open-sourced Muse Glimmer 30B vs Muse Spark 1.2 Glimmer delivered 5 runnable games at roughly one-fifth Muse Spark's cost. Very Interesting experiment by @aimlapi . the setup: • Muse Glimmer 30B — via aimlapi[.]com. cost: $0.02 • Muse Spark 1.2 — via ai
Mac 上的划词翻译工具，好用的大多体积不小，Electron 套壳的后台轻松占几百 MB。 MoePeek 用纯 Swift 写的，安装包 5 MB，后台跑着稳定在 50 MB 左右。 选中文字弹出翻译浮窗，不抢焦点，还支持截图识别翻译和剪贴板翻译。 GitHub： http:// github.com/cosZone/MoePeek
Welcome to the first episode of the Open Model Series with @mervenoyann , @huggingface AI/ML engineer & “open-sourceress”! You’ve probably heard about how you can run “open (source/weight) models” on your own computer – even on your own laptops. We’re her
I'm releasing Roomform, an open-source alternative to the RoomPlan API. It ingests point cloud scans and extracts geometry including walls, doors, windows, objects. No rectangular-room or Manhattan layout assumptions. The model also infers structure behind occ
Fan Bing has open-sourced a new book of over 200,000 words for free. 《Frontline Deployment Engineer》 It’s about how, in the AI era, to weld together “what the product can do” and “what the customer needs”—FDE. The entire book is available to read online, witho
亲测真的强，天下苦 SD2 久矣，终于来了个能打的，还是开源的价格比 SD2 便宜，起步就是 2k 分辨率，AI 视频迎来了真正的 Deepseek 时刻。
its not in the github, its in the brain. github profile matters less now because AI writes the code. irrespective of whether one has a github profile or not, while talking about a problem for 30 mins, people can figure out a person's skill level with a fairly
These open-source drops aren't about releasing local models meant to be ran on consumer compute. They're behemoths that force closed-model AI companies to keep competitive pricing, usage limits, and not gatekeep their best LLMs. It's a diplomatic play. Not onl
google just released 15 AI tools that are completely FREE and can save thousands of $$$ every single monthly. all open-source. MIT licensed. save this in your bookmark." pomelli ( http:// labs.google/pomelli) builds your entire brand identity from just your we
Jensen Huang responds to "Should open-source models be allowed to distill closed-source models?": Distillation is essentially learning, and learning is the foundation of intelligence. Humans have always learned from each other, and AI does the same. Early mode
Governments pay millions for this kind of system. Someone just open-sourced it for free. It's called Crucix. It watches the entire world and texts you when something changes. Pulls from 26 live sources every 15 minutes: - Satellite fire detection, flight and v
Brothers, the "Story-to-Hand-Drawn Video" skill teased last week is officially open-sourced today. Give it a Chinese story, and it can automatically turn it into a hand-drawn diary animation that slowly grows colors from black and white. You can directly input
Someone has open-sourced a director's toolkit on GitHub, specially crafted for Seedance 2.0, completely free of charge Ready-to-use prompt templates covering all sorts of creative starting points From vague ideas and ready-made scripts to image references and
A DEVELOPER CONNECTED CLAUDE TO OBSIDIAN AND TURNED 30+ RANDOM GITHUB REPOS INTO A SEARCHABLE VAULT THAT FINDS DUPLICATES, TRACKS WHAT HE ACTUALLY USES, AND WARNS HIM BEFORE IMPORTANT TOOLS QUIETLY DIE He had the same problem most developers have: star a repo,
THIS REPO JUST SHATTERED THE PAYWALL FOR AI VIDEO GENERATION one developer just completely destroyed Higgsfield with a free, open-source alternative. it is called Open Generative AI. instead of paying monthly subscriptions, you just plug in your own API keys a
CLAUDE CODE TURNED A LANDSCAPING LOGO INTO A 15-SECOND VIDEO AD This guy opened a GitHub repo called arcads-claude-code, copied the URL into Claude Code, and typed: “Please help me install this repo.” Then he uploaded a Summit View Landscape logo and asked Cla
Meta open-sourced a brain-to-text system that reaches 78% word accuracy without surgery. Brain2Qwerty v2 converts non-invasive brain recordings into text with 61% average word accuracy and 78% for its strongest participant. The system reads MEG signals from a
How to manage and monitor a cluster with over 10,000 GPUs? The Tencent team has open-sourced a solution called ARGUS—impressive! Training large models is extremely expensive; for a 10,000-GPU cluster, a single day's electricity and hardware depreciation could
GLM 5.2 JUST BEAT CLAUDE ON REAL BUILDS And the wild part? It’s open-source. The Test: → GLM 5.2 built the best game → Claude Opus 4.8 crushed the orbit map → GPT 5.5 built a playable Color Chain game from one prompt What Actually Mattered: ✓ GLM 5.2 won most
Sam Altman thinks the first one-person billion-dollar company already exists. The next one is a one-person $100,000,000,000 company. Spoiler: the person who builds it is making their own ChatGPT from scratch. At home. Andrej Karpathy put the entire pipeline on
embedded the AI Engineer World Fair talks to be able to “find similar talks”, but then added a map & ability to bookmark a list and export it https:// yoheinakajima.github.io/aie-talks/ cc @swyx
You can build an AI second brain this weekend. No $1000 course. No coding bootcamp. Just your own notes and one free, open-source tool. Here's the whole setup: Step 1 - Put your notes in one folder. Obsidian, markdown, PDFs, old docs. everything, one place. St
Our paper, “Rethinking Psychometric Evaluation of LLMs: When and Why Self-Reports Predict Behavior,” has been selected for Oral Presentation at CTB @icmlconf * Paper: https:// arxiv.org/abs/2606.12730 * Website: https:// psychology-of-ai.github.io/psychometric
“The #3 closed-source LLMs is most in trouble. As enterprises adopt model routing, they're increasingly choosing between the top proprietary models and rapidly improving open-source alternatives. That leaves the number three closed-source provider squeezed fro
today we're launching @Palmier_io , a video editor Claude can edit. use AI to edit, organize, and generate footage directly in the timeline. finally, a video editor built for AI. open-source. mac native. available now.
No mouse. No keyboard. No touchscreen. He controls his computer by pinching the air. Built in a weekend. With Claude. By one person. The hand tracking tech? Free on GitHub. The 3D engine? Free. The AI that wrote all the code? $20/month. In 2002 this was a $100