Polyfire – Javascript SDK to build AI apps without a backend
Victor, Lancelot and Kevin here - we are building Polyfire, it allows you to build AI apps in your frontend without having to worry about deploying any backend or infrastructure, it’s a Firebase style product but for AI apps. Right now, it’s a bit like Vercel AI + LangChain + Pinecone in one Javascript SDK. The repo is https://github.com/polyfire-ai/polyfire-js, our home page is https://polyfire.com. Last June we were working on AI-generated docs but we didn’t quite understand how to make it work. So in July, we decided to open-source everything and focus on the…
In plain words
Polyfire is a JavaScript SDK that enables developers to build AI applications directly in the frontend without managing backend infrastructure or deployment. It combines language models, vector storage, and AI orchestration capabilities in a single library, functioning as an all-in-one solution similar to Firebase but purpose-built for AI apps. The tool is designed for simplicity, requiring minimal setup code to integrate AI features into web applications.
written from the facts on this page · September 2026
From the sources
In the maker’s words, at launch
Victor, Lancelot and Kevin here - we are building Polyfire, it allows you to build AI apps in your frontend without having to worry about deploying any backend or infrastructure, it’s a Firebase style product but for AI apps. Right now, it’s a bit like Vercel AI + LangChain + Pinecone in one Javascript SDK. The repo is https://github.com/polyfire-ai/polyfire-js, our home page is https://polyfire.com. Last June we were working on AI-generated docs but we didn’t quite understand how to make it work. So in July, we decided to open-source everything and focus on the underlying infrastructure. Before this startup, I built at least 20 different apps with Firebase. So I thought it could be really cool to build something like Firebase but to make AI apps. Therefore, Polyfire’s goal is to be simple. Setup takes a couple of lines of code, and then you can call text and image models from your Javascript frontend. It also includes a vector store so you can easily add semantic context to your calls with embeddings. We have more things in the SDK Library like a Chat abstraction with automatic long term memory, DataLoader (e.g. to load text or audio files) and a system to turn prompt in environment variables. We tried to detail as much as possible in our docs: https://docs.polyfire.com. We want to add many more things, your feature requests are welcomed! Our goal right now is to make the best tool to build projects during hackathons: making it super easy to build and experiment with LLMs by adding more integrations, models, and features so hackers have many options. We think we can make something great if we can integrate in one experience the top 10-15 tools people need building AI apps. Give it a look: https://github.com/polyfire-ai/polyfire-js. Let us know what you think!
More ai this month
the category →
I trained a 125M-parameter transformer to autocomplete piano performances in real time (~108 notes/sec on an iPhone 15). The idea is basically GitHub Copilot or Tabnine, except instead of prompting it with code, you prompt it by playing a few notes on a MIDI piano. The model then continues what you played, entirely on-device. The app is free if anyone wants to try it. Happy to answer questions about the model, training, Core ML, or the many things that didn't work.
AI · 17d ago · simedw.com
Astute▲585Automate your B2B brand going viral, with new media creators
AI · 18d ago · company-app.joinastute.com
Hey HN, Henry from Cactus here! We previously released Cactus Needle, a 14MB agentic LLM for tool call, device use, and structured extraction for phones, wearables, smart homes, small robots and microcontrollers. We got really great feedback here, and have now incorporated the suggestions to release Needle 2. The whole model is a single 14MB binary that runs a full session in 28MB of RAM; 45m parameters at 2bit compression. Needle hits 500 tokens/sec decode speed on a Raspberry Pi 5, sits between 400-1,500 tokens/sec on VR devices like Meta Quest 3S and Apple Vision Pro, and ranges…
AI · 26d ago · cactuscompute.com
Launched alongside, October 2023
the whole month →
Nudge 2.0▲1,051In-app experiences to activate, retain, & understand users
Growth · 2023 · nudgenow.com
Unlock AI magic for elevated customer engagement, fast
AI · 2023 · tiledesk.com
- OD
Effortlessly discover API behaviour with a Chrome extension that automatically generates OpenAPI specifications in real time for any app or website.
Dev tools · 2023 · github.com