[태그:] AI
-
Meta Releases Muse Spark Open, a Massive Open-Weight AI Model
Meta’s Muse Spark Open is a new open-weight AI model with 816B parameters that can run locally on two Apple M5 Ultra Mac Studios.
-
Liquid AI’s LFM2.5-24B-A2B Targets Fast On-Device Agents
Liquid AI released LFM2.5-24B-A2B, a 24B-parameter reasoning model with only 2B active parameters, built for fast, private on-device agents.
-
TypeSafe Ships Jev 1.14.0 With Experimental “Danger Zone” Features
TypeSafe’s Jev 1.14.0 adds multimodal images, a new bounding-box type, and longer context via opt-in “danger zone” API flags.
-
OpenCode 2.0 Rebuilds Its AI Coding Agent Around a Shared Server
OpenCode 2.0 overhauls its AI coding agent with a shared background server, a rebuilt API, codemode scripting, and background subagents.
-
Google’s Gemini 4 Argon Targets Long-Horizon Coding and Science
Google DeepMind’s Gemini 4 Argon model tackles long-horizon software engineering, cyber defense, and scientific research, from Chrome refactors to drug discovery tools.
-
Veteran Ruby Developer Ryan Bigg Quits Tech After Dozens of Rejections
Ryan Bigg, a 20-year Ruby developer, is leaving tech after 36 job applications and 24 rejections, citing a brutal job market and AI hiring pressure.
-
Sakana AI’s MASS Lets Models Teach Themselves via Multi-Agent Teams
Sakana AI’s MASS framework lets an AI model optimize, evaluate, and train itself using multi-agent teamwork, a step toward recursive self-improvement.
-
Google Quietly Embraces Device Fingerprinting as Cookies Fade
Google now allows device fingerprinting in its ad products, a tracking method harder to block than cookies. Here is how it works and why it matters.
-
Lightbulb Computer: A Projector-Based Alternative to Smart Glasses
Designer Guillaume Ardaud’s Lightbulb Computer concept uses projection and computer vision to bring ambient, screen-free computing into everyday rooms.
-
Build Your Own Decision Model: Single-Pass LLM Inference
A developer’s guide to building constrained “decision models” that answer multiple-choice questions in one LLM pass, plus how to calibrate their confidence.