Beyond LoRA: Hugging Face Tests Better Fine-Tuning Methods
Hugging Face benchmarked LoRA against newer PEFT methods like OFT, which beat it on image generation while using less memory. The default is not always the best fit.
Blog
Deep dives, tutorials, and analysis for AI-powered creators.
Runway Aleph 2.0 Arrives in Figma Weave
Runway's Aleph 2.0 video model is now available inside Figma Weave, letting designers and editors direct clips up to 30 seconds frame by frame from Weave's node-based canvas.
Cursor Automations Adds /automate and Slack Triggers
Cursor 3.8 expands Automations with a new /automate skill, five new GitHub triggers, and Slack emoji starts for always-on cloud agents.
Gemini Image API Preview Shuts Down June 25: Migrate
Google is shutting down the preview versions of its Gemini image generation models on June 25, 2026. If a tool you rely on still calls a preview model ID, image generation breaks. The fix is a one-line swap to the GA models.
GitHub Copilot Auto-Routes Models to Cut Token Cost
GitHub Copilot now routes each task to the best available model automatically and caches repeated prompt context to cut token cost.
Claude Design Now Imports Your Design System
Anthropic shipped a major Claude Design update on June 17, 2026 that imports your design system from GitHub, edits on a visual canvas, and hands off to Claude Code.
VS Code 1.125 Adds One-Click AI Model Providers
VS Code 1.125 adds one-click model provider installs, a live Copilot budget meter, and managed AI policies for teams.
DeepL Acquires Mixhalo for Live Voice Translation
DeepL has acquired Mixhalo, the real-time audio platform built for concerts and conferences, pushing AI voice translation into live events.
NVIDIA XR AI: Open-Source AI Agents for AR Glasses
NVIDIA released XR AI on June 16, 2026, an open-source library for building AI agents on AR glasses and XR headsets, now in public beta.
Luma Skills: Save and Reuse Creative Workflows
Luma added Skills to Luma Agents, a way to save a creative workflow once and re-run it on new footage or images for consistent results.
Midjourney Draft Mode: 24 Cheap Previews on V8.1
Midjourney added a Draft Mode to V8.1 that generates 24 low-resolution images for about half the fast-hour cost of a standard job.
Cursor Origin: AI-Native Git Forge to Rival GitHub
Cursor announced Origin, a Git-compatible forge built for AI agent workloads. It is a direct shot at GitHub, with general availability set for fall 2026.
SpaceX Buys Cursor for $60B in All-Stock AI Coding Deal
SpaceX is acquiring Cursor, the AI coding tool built by Anysphere, in an all-stock deal valuing it at $60 billion, confirmed June 16, 2026.
Android 17 Adds Gemini Omni Video Editing for Creators
Google shipped Android 17 on June 16, 2026 with Gemini Omni conversational video editing and Screen Reactions, a built-in way to record selfie reactions over your screen.
UCP-Local: Offline RAG for Claude Desktop and Cursor
A new open-source tool, UCP-Local, grounds Claude Desktop, Cursor, and LM Studio in your own files for fully offline retrieval. Released June 16, 2026 under Apache-2.0.
Hammermind: AI Pixel Art and Audio From One Prompt
Hammermind launched on Hacker News as a single-prompt asset pipeline for game developers, turning a text description into game-ready pixel art and audio.