Ollama (chat or unspecified)
How vibe coders and indie builders use Ollama (chat or unspecified): 23 projects built with it since May 2026.
Projects built with it
23
Share of stated stacks, 4 wks
<0.1%
Last 4 weeks vs 8 before
too few to tell
Add-ons built for it
0
Share of stated stacks, per week
Share rather than counts, so changes in how many posts a subreddit lets through don't show up as trends.
Often used with
- Python 6
- Claude Code 5
- Ollama 3
- Qwen (chat or unspecified) 3
- React 2
- GLM (chat or unspecified) 2
- Codex 2
- ChromaDB 2
What's built with it
Latest projects built with Ollama (chat or unspecified)
- Mu3Lab — Bundles self-hosted productivity apps behind a shared AI agent layer where a master agent drives sub-agents that use each app's MCP interface.“Am I making turbo-slopware?”
- Samurai Scribe — Transcribes meetings locally with speaker identification and generates customizable AI summaries saved as Markdown files.“I built a meeting transcription + note-taking app for my mom. 11 months later, it's ready for the world 😌”
- Local RAG Powered Product Review Expert — Answers questions about product reviews by retrieving relevant passages from a local vector database and generating responses with a local language model.“GitHub - Local RAG application using LangChain, Chroma DB, and Ollama.”
- Proto — Runs a local Ollama model that reads Discord messages in a controlled Chrome window and types its replies, with optional Google and YouTube lookups gated by user approval.“Proto: A discord bot that can actually type for you”
- Sometimes the best apps are the… — Lets players stack pancakes in a speeding game with curveballs, bonus pancakes, and helper items that get harder as the stack grows.“Sometimes the best apps are the simplest ones.”
- Test my program — Generates a user experience (images, music, narration, and HTML output) from a single button press using locally run AI models.“Test my program”
- freepress.today — Aggregates links to paywall-free news articles from RSS feeds and APIs, filtering out paywalled sites and publishing them as a static page with hourly updates.“News with paywall-free links only”
- Generar — Generates and edits images locally on an NVIDIA GPU using downloaded open image models, LoRAs and LyCORIS variants, without watermarks or cloud content filters.“Generar - Mejorar Imagenes ( con IA local usando Ollama, python, vue3, fastAPI )”
- fantasy-studio — Turns a text prompt into a playable three.js game by extracting a spec with a local LLM and generating 3D assets on demand.“My GPU died, so I built a prompt→playable-game pipeline that runs entirely on CPU (Ollama + SDXL + image-to-3D + three.js)”
- NexsOS1 — Builds a multi-architecture operating system with a graphical environment, ported applications such as Doom and a text editor, written without library dependencies.“I created a real operating system that runs only on AI”
- PortableMind — Sells a USB stick that runs a full local AI stack with no installation required.“Year one solo: 4,000+ offline-AI USB sticks shipped, no funding, no team. Here's what I got wrong.”
- CouncilAI — Runs multiple local AI models on Windows and routes each question to a model, using parallel answers and voting for harder questions.“I built a Windows AI app that runs 3 local models in a "council" — they debate your question and vote on the best answer”
- RouteStay — Lists accommodation within 25 km of the start or finish of each Tour de France stage, with distances shown.“[Launch] RouteStay.co — find a place to stay near any Tour de France stage”
- Sinain — Watches the macOS screen locally and spawns an AI agent pre-loaded with the context of what the user is stuck on.“Update: I shipped Sinain here a few weeks ago (ambient intelligence for macOS) — it now spots what you’re stuck on and spawns an agent with full context in one click (free, has a local mode)”
- Rag Using Ollama — Answers questions about the contents of an Excel sheet using a local model served by Ollama.“Rag Using Ollama”