kreuzberg

"The decision isn't about which languages to support; it's about what to build with the structured output." tree-sitter-langiage-pack GitHub: github.com/kreuzberg-dev/…

In our newest article, learn why plain text chunking fails code-aware AI agents, how AST-aware chunking fixes it, and how one dependency can replace your entire parser infrastructure. @kreuzberg/why-ai-agents-need-structured-code-intelligence-and-how-to-stop-managing-parsers-4b59a44d5dc0" target="_blank" rel="nofollow noopener">medium.com/@kreuzberg/why…

Introducing Alef⚡️ You write a Rust library; Alef makes it usable in 16 languages with one command. Python, Node, Go, Ruby, Java, C#, PHP, Elixir, WASM, R, Kotlin, Gleam, Zig, C, Swift, Dart. It handles the full pipeline. No manual bindings or glue code. github.com/kreuzberg-dev/…

kreuzberg-txtai is live 🎉 Drop-in replacement for txtai's Textractor. Swap Apache Tika + Java for Kreuzberg's Rust-powered extraction-wide range of formats, stable metadata, zero JVM. pip install kreuzberg-txtai → github.com/kreuzberg-dev/…

Introducing Kreuzcrawl, our high-performance web crawling engine. Built for AI agents from day one, with MCP server integration, real-time streaming, batch operations, and browser rendering for JS-heavy SPAs. 11 language bindings. One core engine.🚀 github.com/kreuzberg-dev/…

@springcentral The kreuzberg-spring-ai DocumentReader handles over 100 formats, has built-in OCR for more than 80 languages, keeps headings when splitting, lets you break content down by elements, and provides detailed metadata. Everything runs locally.

We’ve just released our @springcentral AI integration for Kreuzberg.🎉Get started here github.com/kreuzberg-dev/…

Join the waitlist for Kreuzberg Cloud here kreuzberg.dev

In this article, learn why agentic AI raises the stakes on document quality, what data readiness requires at scale, and how Kreuzberg Cloud will fill this infrastructure gap💡@kreuzberg/beyond-the-model-why-document-intelligence-is-the-next-ai-infrastructure-layer-3ca0a7d18fb9?postPublishedType=repub" target="_blank" rel="nofollow noopener">medium.com/@kreuzberg/bey…

🔴 Live now: twitch.tv/namihirschfeld Want to see how @kreuzberg_dev gets built? Now's your chance. Our co-founder is streaming live- come, ask questions, and watch content intelligence take shape in real time. We'll be doing this a few times a week, so bookmark this channel ;)

Our tree-sitter-language-pack (v1.6) now supports 305 languages 🔥. Agents using it can process source code across 305 languages with the same structured output. No per-language setup required. MIT licensed. Open source. GitHub: github.com/kreuzberg-dev/…

@Haystack_AI Let us know what you think on our Discord server: discord.com/invite/xt9WY3G…

Flawed document extraction is one of the biggest bottlenecks in RAG and most pipelines don't see it coming. Find KreuzbergConverter in @Haystack_AI's core integrations. 91+ formats, local OCR, one component. Read more: @kreuzberg/the-haystack-converter-that-handles-91-file-formats-without-a-cloud-api-0505b51e49fb" target="_blank" rel="nofollow noopener">medium.com/@kreuzberg/the…

@itsafiz @Haystack_AI

@Haystack_AI @kreuzberg_dev you guys are doing really great. I want to give it a try and I will make a post on it soon. Keep going!

Most document parsing pipelines still rely on cloud APIs, external services, or brittle format-specific libraries stitched together. @kreuzberg_dev takes a different approach: a Rust-core document intelligence engine that extracts text, tables, and metadata from 91+ file formats entirely locally. No API calls. No data leaving your infrastructure. We've now integrated it into Haystack as a converter component. Drop in KreuzbergConverter to transform PDFs, DOCX, PPTX, scanned images, emails, archives, notebooks, and more into Haystack Document objects. 🐍 pip install kreuzberg-haystack 🔗 Documentation: haystack.deepset.ai/integrations/k…

KreuzbergConverter sits at the entry point of any indexing pipeline, turning raw files into clean Haystack Documents. Tables come out as structured output, languages are detected automatically, and each document carries a quality score. pip install kreuzberg-haystack 💥

@Haystack_AI Awesome, thank you for the shoutout! Excited to see Kreuzberg in the Haystack ecosystem- local-first document intelligence across the full format breadth.

KreuzbergConverter is now part of Haystack's core integrations and is managed upstream by deepset, makers of Haystack: github.com/deepset-ai/hay…

Kreuzberg now has integrations with three of the most widely used frameworks for building AI applications: @llama_index, Haystack by @deepset_ai, and @crewAIInc. No matter what stack you are using, you can easily connect Kreuzberg's document intelligence engine with them.

If you're building RAG pipelines with LlamaIndex, two new packages give you structure-aware document ingestion out of the box. Try it out on GitHub: github.com/kreuzberg-dev/…

@llama_index @deepset_ai @crewAIInc @crewAIInc

@llama_index @deepset_ai @crewAIInc CrewAI agents are designed to reason and collaborate, but they can't read files without additional support. kreuzberg-crewai solves this problem. github.com/kreuzberg-dev/…