Generative AI news and analysis on Web Pulse: 15457 curated articles tracking the latest Generative AI developments, tools, and research, updated continuously from vetted sources.
A developer built a benchmark testing whether a general-purpose LLM can infer fraud from transaction-graph context rendered as text, using the IEEE-CIS Fraud Detection dataset. Gemma 4 31B (google/gemma-4-31b) scored 55%…
A developer built WildQuest, an open-source AI-powered outdoor adventure generator that turns a simple walk into an interactive quest. The project uses a React frontend and a Python FastAPI backend, with Google's Gemini …
A developer built Sidequestmaxxing, an open-source AI application that turns a neighborhood into an open-world game map, using a small open-weight language model fine-tuned as an AI Dungeon Master to generate one real-wo…
A developer spent several months building an AI interior design app after finding that image-to-image models regenerate rooms rather than redesign them, moving windows and erasing architectural features. The fix was a de…
A developer built ChromaWild, a local color-walk and field-journal app that uses Gemma 2B through Ollama to generate coherent color-and-subject quests for outdoor observation. The Next.js API route prompts the model for …
A developer built Golden Hour, an open-source tool that uses plain Python to score every 20–40 minute outdoor window in a user's free time against Open-Meteo forecast data and then has Gemma 3, served via Ollama, write o…
A developer built GrowWise, an open-source AI gardening companion that runs on a locally hosted open-weight model with no API tokens or cloud AI costs. The app combines an AI Garden Coach, outdoor "Touch Grass" missions,…
A developer built NatureSnap AI, an open-source outdoor exploration app powered by Google's open-weight Gemma 2 and PaliGemma models that identifies plants, insects, rocks and mushrooms from photos and then pushes users …
A developer built a small Pillow and OpenCV script for evaluating outdoor portrait generators, finding that whole-frame Laplacian variance misleads because blurred backgrounds dominate the image. On a calibration dayligh…
A developer built TrailShield, a phone-only web app that uses Google's open-weight Gemma model (gemma-4-26b-a4b-it) to give short, cautious notes on trail photos and then prompts users to put their phone away. The app st…
A developer built Loot Goblin, a web app that turns photos of real-world objects into AI-generated RPG loot with rarity, stats, lore, and a Goblin Coin value stored in a persistent Treasury. The project uses React, Vite,…
A developer built a 17-agent AI swarm that runs entirely on a Pixel 7 smartphone, using llama-cpp-python and a quantized Phi-2 model to analyze long-form text and generate tailored social media posts for X, LinkedIn, Fac…
A developer built Touch Grass Planner, a local-first web app that uses a locally running Gemma 3 4B model via Ollama to write short outdoor outing plans, then dims the screen to a dark countdown timer with a route map. T…
A developer building a video game describes a hybrid workflow that confines AI agents to auxiliary, engine-facing code while keeping the engine-agnostic /Game/ folder — map generation, NPC behaviour, buildings — off-limi…
A developer built StepOut, an open-source AI tool that generates personalized real-world outdoor missions designed to get users off their screens, using Google's open-weight Gemma model served through Ollama. The project…
A developer built SideQuest, a Next.js app that uses Google's Gemma 4 26B model to generate short outdoor missions and then prompts users to put their phone away, with a Gemini 3.7 Flash verifier checking a single return…
Odyssey released Odyssey-3, a 14-billion-parameter generative world model, as a free public research preview that generates interactive environments from text prompts in real time and can reportedly control robots, drone…
The European Union formally published final guidelines for its Artificial Intelligence Act, establishing a risk-based framework that targets "systemic risk" providers such as OpenAI, Google, and Meta with fines of up to …
A developer built Seekr, an open-source, map-based treasure hunt app for real-world events in which organizers pin riddles and hidden locations and teams of up to four solve clues, walk to each spot, and verify arrival w…
Alibaba's Qwen team released Qwen-Image-2.1, a roughly 20-billion-parameter multimodal diffusion model that renders legible, correctly spelled text inside generated images across Chinese, English, Japanese and Korean. Th…