SmolCopilot – 360M LLM writing assistant in the browser
Hey! I've been really intrigued about small, on device models recently. This project was to explore that a small base model running in the browser as a Co Pilot-like writing assistant. You can check out the demo here: smol-complete.vercel.app. Smal, on device models have a lot of advantages with low latency, cost, privacy (a big one), and being task specific. The demo will definitely repeat itself and hallucinate sometimes, but let me know what you think!
SmolCopilot – 360M LLM writing assistant in the browser launched on August 20, 2024 with 11 votes, #1,388 of 5,471 launches that month and more than 75% of that year's launches. The site was still online in September 2026, 2 years on.
In plain words
SmolCopilot is a writing assistant powered by a 360M language model that runs directly in the browser. It provides real-time writing suggestions with low latency while keeping data private on the user's device. The tool is designed for writers and content creators who want AI assistance without sending text to external servers. Being lightweight and task-specific, it prioritizes speed and privacy over larger cloud-based models.
written from the facts on this page · September 2026
From the sources
360M model running in the browser on WebGPU. Contribute to cfahlgren1/smol-copilot development by creating an account on GitHub.
Does a similar job
all alternatives →
Smol machines – subsecond coldstart, portable virtual machinesApr 2026 · github.com · ▲499Branchable computing by using a portable, lightweight, self-contained virtual machine - smol-machines/smolvm
Z80-μLM, a 'Conversational AI' That Fits in 40KBDec 2025 · github.com · ▲514How small can a language model be while still doing something useful? I wanted to find out, and had some spare time over the holidays. Z80-μLM is a character-level language model with 2-bit quantized weights ({-2,-1,0,+1}) that runs on a Z80 with 64KB RAM. The entire thing: inference, weights, chat UI, it all fits in a 40KB .COM file that you can run in a CP/M emulator and hopefully even real hardware! It won't write your emails, but it can be trained to play a stripped down version of 20 Questions, and is sometimes able to maintain the illusion of having simple but terse conversations…
Smolmodels – open-source tool to build ML models using natural language2025 · github.com · ▲37Hi HN - Marcello and Vaibhav here. We built smolmodels to experiment with using LLMs for ML development. It's a fully open-source library that generates complete model training and inference code from natural language descriptions. It combines graph search with LLM code generation to find a model that gives as good predictions as possible. The core idea is that LLMs are overkill for a lot of predictive tasks. Smolmodels automates the trial-and-error process of finding the right model architecture and training approach, letting you build small, specialised models. You can either provide your…
Rapidpages – OSS alternative to vercel's v02023 · github.com · ▲216Hey everyone, Really excited to share what I've been working on. Rapidpages is a prompt-first online IDE, think midjourney for front-end developers. I've been working on this for a while and it's great to see some interest from companies like Vercel in this space. All you need for self-hosting is an OpenAI key and a GitHub oauth app. Simply clone the repo and play with it. It's also available on the cloud at www.rapidpages.io Please give it a try and let me know if you have any feedback, and if you like what I'm doing with Rapidpages, please give it a star on GitHub. Thanks!
More ai this month
the category →
Paintings by AI models: programs run through a simulation of oil paint.
AI · 22d ago · stillwet.art
FastRouter.ai▲364Route requests to the right LLM for cost, latency & quality
AI · 23d ago · fastrouter.ai
Source: Show HN post ↗
Launched alongside, August 2024
the whole month →
Jupitrr AI▲1,195Jupitrr AI turns raw talking videos into scroll-stopping content for coaches and consultants. Research, script, record, edit, and publish in one app.
AI · 2024 · jupitrr.com
Hey there HN! We’re Joe and Stopa, and today we’re open sourcing InstantDB, a client-side database that makes it easy to build real-time and collaborative apps like Notion and Figma. Building modern apps these days involves a lot of schleps. For a basic CRUD app you need to spin up servers, wire up endpoints, integrate auth, add permissions, and then marshal data from the backend to the frontend and back again. If you want to deliver a buttery smooth user experience, you’ll need to add optimistic updates and rollbacks. We do these steps over and over for every feature we build, which can…
Dev tools · 2024 · github.com