{"id":90330,"title":"Chonkie 🦛 Open Source Data Ingestion for AI","tagline":"Feed your models better. Chonkie cleans, chunks, and makes your data AI ready","body":"![uploaded image](/media/?type=post\u0026id=90330\u0026key=user_uploads/1849274/669a375f-f24a-4d4b-9cca-c668ce5d3da4)\n\n## 🚀 TL;DR\n\nModels are now a commodity. The real differentiator for AI applications is _how effectively they use their data_. However, building robust data systems can be both complex and time-consuming.\n\nChonkie fixes this. We make it easy to bring high-quality data ingestion and context-building to your AI projects, ensuring your application is accurate, fast, and cost-efficient.\n\nWe’re open source on GitHub! ([Python](https://github.com/chonkie-inc/chonkie), [TypeScript](https://github.com/chonkie-inc/chonkie-ts)). Leave us a ⭐!\n\n# 🎬 Launch Video\n\nhttps://www.youtube.com/watch?v=eBvCCEgbNDw\n\n---\n\n---\n\n---\n\n---\n\n---\n\n---\n\n---\n\n---\n\n## 👀 Are you…\n\n👷 **A Builder**\n\n* Install with [pip install chonkie ](https://pypi.org/project/chonkie/)or [npm i chonkie](https://www.npmjs.com/package/chonkie)\n* Use it for free in your projects\n* Join our [Discord](https://discord.gg/6V5pqvqsCY) for direct support from the founders!\n\n🏢 **Business Looking To Go AI Native**\n\n* Want a managed ingestion pipeline? [Contact us](https://cal.com/shreyashn/chonkie-demo)\n* Enjoy a white-glove onboarding experience\n* Use our hosted solution or deploy on-prem!\n\n## ✨ Why Chonkie?\n\nMost AI products today don’t fail because of the model. They fail because they’re fed disorganized, bloated, or incomplete data. The results in wrong answers, slow performance, and expensive bills. There are some existing tools that try to address this, however they are either:\n\n* 🔻 Too slow or bloated\n* 🧱 Too barebones — feels like reinventing the wheel every time\n\nSo we built **Chonkie** — fast, lightweight, and comprehensive.\n\n**Chonkie + Your Amazing AI Application =**\n\n* 🎯 More accurate result\n* ⚡ Faster response times\n* 💸 Lower token usage\n\n## 🔧 What Can Chonkie Do?\n\nChonkie handles the full data ingestion pipeline for AI projects:\n\n* 📄**Bring Any Document**: All formats welcome\n* 🍳**Clean It with Our Chefs**: Normalize and prep your data\n* 🧩**Chunk It**: Break documents into meaningful pieces\n* 🛠️**Refine Your Chunks:** Add context, metadata, or convert to embeddings\n* 🔌**Connect Anywhere**: Integrate with any database and fetch your chunks with ease\n\n## 🎥 See It in Action\n\nSee how Chonkie splits your data, reduces token costs by over **75%**, and generates citations automatically!\n\n![uploaded image](/media/?type=post\u0026id=90330\u0026key=user_uploads/1849274/5ec9dbe7-37c8-428d-b589-8d71c6195191)\n\n## ✅ Try Chonkie Today\n\n* 📬Looking for a managed enterprise solution? [Contact us](https://cal.com/shreyashn/chonkie-demo)\n* 💻 Install our open source offering with  pip install chonkie or npm i chonkie\n* 💬Join our [Discord](https://discord.gg/6V5pqvqsCY) for support and community\n\n\\~Happy Chonking! 🦛✨","slug":"NUw-chonkie-open-source-data-ingestion-for-ai","created_at":"2025-05-13T13:34:25.524Z","updated_at":"2026-07-22T10:44:36.242Z","total_vote_count":80,"url":"https://www.ycombinator.com/launches/NUw-chonkie-open-source-data-ingestion-for-ai","share_image_url":"https://www.ycombinator.com/media/?type=post\u0026id=90330\u0026key=user_uploads/1849274/669a375f-f24a-4d4b-9cca-c668ce5d3da4","company":{"id":30431,"name":"Feyn","slug":"feyn","url":"https://www.usefeyn.com","logo":"https://bookface-images.s3.amazonaws.com/small_logos/7711f83eeb988737aadd19d8943c753d2e4085f4.png","batch":"Spring 2025","industry":"B2B","tags":["Artificial Intelligence","Developer Tools","AI","ML"],"search_path":"https://bookface.ycombinator.com/company/30431"}}