{"slug": "cityscape-with-flying-cars", "title": "cityscape with flying cars", "summary": "A developer built Aether Grid, an open-source real-time 3D futuristic cityscape in which a locally run Llama 3.2 3B model, served through Ollama, acts as the traffic controller instead of hardcoded rules or a physics engine. The model returns structured JSON — spawn, reroute and atmosphere fields — that a parser maps to Three.js scene actions, letting users type natural-language commands like \"a storm is rolling in from the east\" to make vehicles scatter and the sky darken. The project runs entirely on-device at roughly 15 tokens per second, with no API or cloud calls, and is MIT-licensed under the Hacktoberfest 2026 tag.", "body_md": "I Built a Living Futuristic City Where Open-Source AI Drives the Traffic\n\nTags: #hf26challenge #hacktoberfest2026 #ai #webdev #open-source\n\nWhat I Built\n\nAether Grid is a real-time, procedurally generated futuristic cityscape where every flying car, drone taxi, and sky-rail is orchestrated by a local open-weight language model. No API calls. No cloud. The city thinks for itself.\n\nThe core idea: instead of hand-coding traffic patterns or using a physics engine, a small open-weight model (running via Ollama) acts as the city's nervous system. It decides:\n\nYou get a 3D city rendered with Three.js that responds to natural-language commands. Type \"a storm is rolling in from the east\" and the vehicles scatter, neon signs flicker, and the sky darkens — all decided by the model, not by hardcoded rules.\n\nDemo\n\n[Video / Live Demo Link Here]\n\nCode\n\n[GitHub Repository Link Here]\n\nRepo structure:\n\naether-grid/\n\n├── src/\n\n│   ├── city/          # Three.js scene, buildings, roads\n\n│   ├── traffic/       # Vehicle spawning, lane logic\n\n│   ├── ai/\n\n│   │   ├── prompt.js  # System prompt for the city model\n\n│   │   ├── parse.js   # JSON response → scene actions\n\n│   │   └── client.js  # Ollama local inference wrapper\n\n│   └── ui/            # Prompt input, stats overlay\n\n├── models/            # Local model config (llama3.2:3b)\n\n└── README.md\n\nHow I Built It\n\nThe AI layer. I used Llama 3.2 3B via Ollama, running entirely on my machine. The system prompt instructs the model to act as a city traffic controller and return structured JSON:\n\n{\n\n  \"spawn\": [{ \"type\": \"taxi\", \"lane\": 3, \"speed\": 1.2 }],\n\n  \"reroute\": [{ \"vehicle_id\": \"v-42\", \"new_lane\": 7 }],\n\n  \"atmosphere\": { \"sky\": \"storm\", \"lighting\": \"neon-amber\" }\n\n}\n\nThe response parser maps each field to a Three.js action. No external API — the model runs at ~15 tokens/sec on my laptop, which is plenty for a city that updates every 2 seconds.\n\nThe 3D scene. Buildings are instanced meshes with emissive window textures. Flying cars are simple box-and-wing geometries on spline paths. The trick that made it feel alive: each vehicle has a slight sinusoidal bob, and the camera slowly orbits the city center unless the user takes control.\n\nThe prompt interface. A single text input at the bottom. You describe what's happening in the city, the model decides how the city reacts. This is where the open-weight model does the heavy lifting — it's not generating text for a user, it's generating actions for a simulation.\n\nWhy Open Innovation Matters\n\nThis is the part I care about most.\n\nI could have called a hosted API to do this. I didn't, and here's why:\n\nThat last point is what excites me about open-weight models. The barrier between \"I have an idea for how a city should feel\" and \"here's a city that feels that way\" is a text file.\n\nWhat's Next\n\nBuilt during Hacktoberfest 2026. All code is MIT-licensed. Go build a city.", "url": "https://wpnews.pro/news/cityscape-with-flying-cars", "canonical_source": "https://dev.to/insinct-coder/cityscape-with-flying-cars-41bf", "published_at": "2026-11-02 00:12:13+00:00", "updated_at": "2026-11-02 00:22:34.305560+00:00", "lang": "en", "topics": ["artificial-intelligence", "large-language-models", "ai-agents", "generative-ai"], "entities": ["Aether Grid", "Llama 3.2 3B", "Ollama", "Three.js", "Hacktoberfest 2026"], "also_reported_by": [], "alternates": {"html": "https://wpnews.pro/news/cityscape-with-flying-cars", "markdown": "https://wpnews.pro/news/cityscape-with-flying-cars.md", "text": "https://wpnews.pro/news/cityscape-with-flying-cars.txt", "jsonld": "https://wpnews.pro/news/cityscape-with-flying-cars.jsonld"}}