Kirk Marple

Graphlit — API for LLM enabled knowledge ingestion and retrieval

Featured
10
•
Graphlit simplifies the development of intelligent AI apps and services. Compared to LangChain or LlamaIndex, which require the DIY combo of vector dbs, parsers and data pipelines, our managed platform lets you focus on building your app - not your AI infra.

Add a comment

Replies
Best
Kirk Marple
Maker
đź“Ś
👋 Hey Product Hunt! I'm happy to formally launch Graphlit on Product Hunt: our API platform for building AI apps. If you're building AI copilots, chatbots or other vertical AI apps, Graphlit simplifies and accelerates your development. Compared to existing Open Source solutions like LangChain or LlamaIndex, which require the DIY combination of vector databases, LLM embeddings, cloud storage and data pipelines, our managed Graphlit platform handles your AI and data infrastructure for you. No need to be limited by the OpenAI Assistants API - no file size limits (on all tiers), or storage limitations (on the Growth tier). 💸 Graphlit is free to use, up to 1GB of content, and paid plans start at $49/mo + credit usage. 💡 Signup today, and you can be ingesting data and having LLM conversations within minutes. With integrated web scraping, Graphlit ingests existing websites by sitemap. With built-in audio transcription, it indexes podcasts, videos and meeting recordings. Any format of unstructured data will be made searchable, via metadata filtering and text and image embeddings. Slack, Notion, Google Mail and Microsoft Outlook email are supported as data feeds. Create automated LLM-generated alerts on people, places, companies or topics found in your content. Use any content with RAG conversations, even images or websites described with the GPT-4 Vision model. ✅ API first: Made for app developers, not data scientists ✅ Graph-based: via LLMs, we build a knowledge graph from your unstructured data ✅ Multi-modal RAG: not just PDFs and web pages, we support audio, video and images ✅ Model-agnostic: we handle prompted retrieval with models from OpenAI, Anthropic, Meta, Mistral, etc. ✅ Managed cloud-native platform: fully automated unstructured data ETL pipelines ✅ No assembly required: no need for Langchain, Pinecone, S3, etc. ✅ Built-in multi-tenancy, semantic search, storage and workflow automation ✍️ Read more about use cases for Graphlit: — Slack audio alerts — AI-generated podcasts — GPT-4 Vision for image analysis — LLM tools for data extraction — Reddit market intelligence with LLM 🔥 Get started with our API documentation. 🆕 See our changelog for all the latest features. We want to learn how we can help you build your AI apps faster with Graphlit. Thanks for trying us out! 👋 - Kirk, Founder/CEO
Anastasia Chakileva
Thrilled to see Graphlit on Product Hunt! As someone keen on AI innovation, I'm fascinated by how it simplifies AI app development, especially compared to the DIY approach of LangChain and LlamaIndex. The seamless handling of diverse data formats – from audio to emails – is a game changer. Curious to know, how does Graphlit ensure data security with such a wide range of inputs?
Kirk Marple
@anastasia_chakileva_ Appreciate your interest! We provision isolated cloud storage for each customer's project, which is also encrypted at rest. And, we provide multi-tenancy so you get access control for the content across your app's users.
Dhirendra Singh
Simplify the development of intelligent AI apps and services with Graphlit's managed platform.
Garen Orchyan
Big congrats on the launch, best of luck🦄❤️
Rick Fan
Congrats, powerful APIs! good luck!
Muhammad Ali abba
Wow... good knowledge and it's very easy to understand
Ghost Kitty
Comment Deleted
Obie
Hi Product Hunt. I'm a happy customer of Graphlit, so very happy to see their launch! Kirk is the absolute best guy, always on top of whatever questions or feedback I have.