Generative AI news and analysis on Web Pulse: 11709 curated articles tracking the latest Generative AI developments, tools, and research, updated continuously from vetted sources.
Fal AI's hosted Hunyuan H3 Max generates 5-second video clips in 2-3 seconds, while local H3 workflows on consumer GPUs take 80-90 seconds per clip but cost nothing beyond electricity, according to a hands-on comparison.…
Hunyuan Video 3, an open-weight video generation model, can now be run locally in ComfyUI with an 8-step Turbo LoRA and sage attention to generate 5-second clips in roughly 80 to 90 seconds on a capable home GPU, down fr…
Amazon Quick and fal have partnered to build agentic creative workflows that address the 78% of creative leaders reporting demand exceeds team capacity, using Amazon Quick as the agent workspace, fal's platform of more t…
Google is expanding Gemini in Chrome into a more context-aware desktop experience with a new Personal Intelligence beta that connects data from Google apps like Gmail, Calendar, and Docs to provide personalized assistanc…
Adobe announced on Thursday a new AI Assisted Editor mode for Photoshop, currently in beta, that lets users edit images with natural language prompts alongside the classic Pro Editor. The feature, part of a broader updat…
A user on the Kagifeedback forum suggests that an AI-generated audio clip containing screaming is normal in gaming and amateur-made videos, and that the AI did nothing wrong by producing it. The user notes that the phras…
Sopro V2, a CPU-optimized voice cloning model, enables high-quality text-to-speech synthesis on standard hardware without GPUs, using small audio samples of a few minutes for training. This development removes hardware b…
Tim O'Reilly discusses the integration of AI into the writing process, sharing his experiences and insights on how AI tools can assist authors. He explores the potential benefits and challenges of AI-assisted writing, em…
Search Engine Land's 2027 SEO roadmap argues that AI-generated answers and zero-click results are redefining search visibility, citing Google I/O 2026 figures of over 1 billion monthly AI Mode users and 2.5 billion AI Ov…
Google released Gemini 3.5 Transcribe, a speech-to-text model supporting real-time transcription across 85+ languages with custom vocabulary recognition, and Gemini Omni 1.1 Flash, a video model with scene extension, fra…
Flow matching, a generative modelling technique, learns a velocity field that transports noise directly onto data in one continuous motion, avoiding the stochastic reversal of diffusion models. The method, detailed in a …
An engineer detailed the architecture of high-volume generative media applications, emphasizing the need for distributed asset management and CDN optimization. The system treats media assets as deterministic projections …
Google launched Expert Intelligence, a cross-Google initiative that lets users add purchased Google Play Books ebooks to Gemini Notebook as sources, with over 100,000 titles from publishers including Bloomsbury, De Gruyt…
Jeff Huleatt, a developer, created an Agent Skill called paper-to-astro-mdx that uses Gemini to transcribe handwritten blog post drafts into Astro-friendly MDX, after being inspired by a Hacker News post about blogging o…
The moeinGTS team, led by Ali and Arshiya Sohrevardi, has released an open-source family of AI models designed for local deployment, including lightweight LLMs (1.5b and 3b), a flagship model (GTS-1), a vision adapter (M…
State Media FM, a free open-source (Apache-2.0) app, turns team activity from GitHub, GitLab, and Hacker News into a private radio station with generative music and spoken bulletins, run locally via a single command. The…
Google is connecting its AI content creation, application development, and agent deployment tools with recent updates to Flow, Google AI Studio, and Gemini Enterprise. Flow now offers scene extension and object insertion…
A developer has spent six months running a one-person film studio using Anthropic's Claude Code, producing a 10-episode series. The setup relies on a CLAUDE.md file that defines Claude's role as a creative director, SKIL…
Google's Gemini Omni 1.1 Flash video model now analyzes up to ten seconds of existing footage instead of just the last second for more consistent scene extensions, and scenes can be extended in 10-second increments up to…