CHOI (@arrakis_ai)

Chrome 안에서 직접 Codex를 사용하는 Chormex v0.1.3이 공개됐다. MCP와 플러그인 연결, 페이지 기반 코믹 생성, 선택한 컨텍스트 질의, 사이트별 제안, Plan Mode 등 브라우저 내 AI 작업을 확장하는 기능이 추가됐다.

https://x.com/arrakis_ai/status/2051196116302225790

#codex #chrome #mcp #browserai #extension

CHOI (@arrakis_ai) on X

Use Codex directly inside Chrome. Chormex just dropped v0.1.3 and it’s kind of wild. You can now connect Skills, plug into MCP and plugins, generate comics from the page you’re on, highlight any context and ask questions instantly, get site-specific suggestions, use Plan Mode,

X (formerly Twitter)
Google Chrome Now Lets You Save AI Prompts as Reusable Skills

Google launched Skills in Chrome, letting users save and rerun Gemini prompt workflows with one click. The feature includes a starter library and confirmation safeguards for sensitive actions.

Nico Martin (@nic_o_martin)

Transformers.js v4.1이 공개되었으며, 이전에는 브라우저에서 실행되지 않던 기능이 한 줄의 코드로 가능해졌다고 예고했다. 브라우저 기반 AI 추론과 경량 모델 활용을 크게 확장할 수 있는 중요한 프런트엔드 AI 도구 업데이트다.

https://x.com/nic_o_martin/status/2044408658327454122

#transformersjs #browserai #javascript #llm #opensource

🤷 Nico Martin (@nic_o_martin) on X

Transformers.js v4.1 is out 🎉 Something that has literally never run in a browser before is now possible. One line of code. Can you guess what it is? 👇 (hint: think small 🤏)

X (formerly Twitter)

Chrome Skills let users save Gemini prompts as shortcuts for repeated tasks like page summaries or product comparisons across browser tabs. Google announced the feature in April 2026, though rollout appears uneven across accounts and regions. The tutorial covers building reusable workflows while avoiding automation pitfalls, but access remains limited during the gradual deployment phase.

#ChromeSkills #Gemini #BrowserAI

https://www.implicator.ai/intermediate-tutorial-build-useful-chrome-skills-for-gemini-workflows/

Intermediate Tutorial: Build Useful Chrome Skills for Gemini Workflows

Chrome Skills turn repeated Gemini in Chrome prompts into reusable browser workflows. This intermediate tutorial shows how to confirm access, save useful prompts, compare selected tabs, build a research verification Skill, and keep the library from becoming clutter. It also explains where the rollout is limited, why selected tabs matter, when a saved prompt is too risky for confidential work, and how to test outputs before trusting them.

Implicator.ai

Chrome (@googlechrome)

Chrome에 곧 새 기능 Skills가 추가된다. 자주 쓰는 Gemini 프롬프트를 저장해 재사용하거나, 미리 만들어진 프롬프트 라이브러리에서 선택할 수 있다. 브라우저와 AI 프롬프트 활용성을 높이는 기능 업데이트다.

https://x.com/googlechrome/status/2044099474348224681

#chrome #gemini #prompting #browserai #productivity

Chrome (@googlechrome) on X

Keep tabs on our latest Chrome feature, coming soon: Skills! Save your go-to Gemini in Chrome prompts to reuse them later, or choose from a library of premade prompts.

X (formerly Twitter)

What if AI runs directly inside the browser?

No external APIs.
No data leaving the device.

Privacy-first AI is becoming real.

Explore browser-based AI at Voxxed Days Zürich.

📅 https://vdz26.voxxeddays.ch/

#BrowserAI #WebDevelopment #VDZ26

Xenova (@xenovacom)

NVIDIA의 새로운 Nemotron-3-Nano(4B) 모델이 소개되었습니다. Hybrid Mamba+Attention 구조를 사용해 추론과 비추론 작업을 하나의 모델로 통합했으며, 매우 작고 효율적이라 웹 브라우저에서 100% 로컬로 초당 75토큰 속도로 실행된다고 합니다.

https://x.com/xenovacom/status/2034748804587508045

#nvidia #nemotron #llm #mamba #browserai

Xenova (@xenovacom) on X

Not enough people are talking about NVIDIA's new Nemotron-3-Nano (4B) model! 🤯 Hybrid Mamba + Attention architecture, designed as a unified model for reasoning and non-reasoning tasks. So small and efficient, it can run 100% locally in your web browser at 75 tokens per second.

X (formerly Twitter)

Amazon’s AGI Lab loses its visionary head, David Luan, after steering the groundbreaking Nova Act that turned browsers into autonomous AI agents. What this means for web automation and the future of artificial intelligence? Read on for the full story. #AGI #NovaAct #AIagents #BrowserAI

🔗 https://aidailypost.com/news/amazon-agi-lab-head-david-luan-departs-after-leading-nova-act

AA (@measure_plan)

사용자가 만든 컴퓨터 비전 시스템이 브라우저 상에서 WebGPU를 활용해 실행된다고 명확히 밝혔다. ML 모델은 Transformers.js를 통해 로드되며, MacBook M2 환경에서는 FPS가 낮지만, 클러스터 환경에서의 실험을 제안했다. 이는 경량 AI 모델의 웹 실시간 추론 구현 가능성을 시사한다.

https://x.com/measure_plan/status/2025998234951266528

#webgpu #transformersjs #browserai #ai #realtime

AA (@measure_plan) on X

@xenovacom my original post wasn't clear, so i want to point out that this runs in the browser using webgpu the ML models are loaded via transformers js fps is pretty slow because of my m2 laptop. if anyone wants to try with a cluster of mac minis / studios, just dm me :]

X (formerly Twitter)