Krira Chunker
Rust-native doc chunker for RAG-40x faster than LangChain
What it does
Most RAG pipelines bottleneck at chunking. LangChain's pure Python chunker is slow and memory-hungry at scale. Krira Chunker fixes this with a Rust core + Python bindings: → 40x faster than LangChain → O(1) memory — flat at any scale → Drop-in Python API → 17 versions shipped, 315+ installs pip install krira-augment Built for devs tired of chunking being their pipeline's weak link.
Does a similar job
all alternatives →- CAChonkie – A Fast, Lightweight Text Chunking Library for RAG2024 · github.com · ▲199
I built Chonkie because I was tired of rewriting chunking code for RAG applications. Existing libraries were either too bloated (80MB+) or too basic, with no middle ground. Core features: - 21MB default install vs 80-171MB alternatives - 33x faster token chunking than popular alternatives - Supports multiple chunking strategies: token, word, sentence, and semantic - Works with all major tokenizers (transformers, tokenizers, tiktoken) - Zero external dependencies for basic functionality Technical optimizations: - Uses tiktoken with multi-threading for faster tokenization - Implements…
- FBFastGraphRAG – Better RAG using good old PageRank2024 · github.com · ▲457
Hey there HN! We’re Antonio, Luca, and Yuhang, and we’re excited to introduce Fast GraphRAG, an open-source RAG approach that leverages knowledge graphs and the 25 years old PageRank for better information retrieval and reasoning. Building a good RAG pipeline these days takes a lot of manual optimizations. Most engineers intuitively start from naive RAG: throw everything in a vector database and hope that semantic search is powerful enough. This can work for use cases where accuracy isn’t too important and hallucinations are tolerable, but it doesn’t work for more difficult queries that…
- WAWhirlwind – Async concurrent hashmap for Rust2024 · github.com · ▲140
Hey HN, this is Will and David from Fortress (https://news.ycombinator.com/item?id=41426998). We use a lot of async Rust internally, and created this library out of a need for an async-aware concurrent hashmap since there weren’t many available in the Rust ecosystem. Whirlwind is a sharded HashMap with a fully asynchronous API. Just as dashmap is a replacement for std::sync::RwLock, whirlwind aims to be a replacement for tokio::sync::RwLock. It has a similar design and performance characteristics to dashmap, but seems to perform better in read-heavy workloads with tokio's…
- AFApache Fory Rust – 10-20x faster serialization than JSON/ProtobufOct 2025 · fory.apache.org · ▲67
Serialization framework with some interesting numbers: 10-20x faster on nested objects than json/protobuf. Technical approach: compile-time codegen (no reflection), compact binary protocol with meta-packing, little-endian layout optimized for modern CPUs. Unique features that other fast serializers don't have: - Cross-language without IDL files (Rust ↔ Python/Java/Go) - Trait object serialization (Box) - Automatic circular reference handling - Schema evolution without coordination Happy to discuss design trade-offs. Benchmarks:…
- PAPython at the Speed of Rust2025 · blog.fxn.ai · ▲68
I’m sure many of you are familiar, but there’s a treacherous gap between finding (or building) a model that works in PyTorch, and getting that deployed into your application, especially in consumer-facing applications. I’ve been very interested in solving this problem with a great developer experience. Over time, I gradually realized that the highest-impact thing to have was a way to go from existing Python code to a self-contained native binary—in other words, a Python compiler. I was already pretty familiar with a successful attempt: when Apple introduced armv8 on the iPhone 5s, they…
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 · 18d ago · simedw.com
Astute▲585Automate your B2B brand going viral, with new media creators
AI · 20d 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 · 28d ago · cactuscompute.com
Source: Product Hunt launch ↗
Launched alongside, March 2026
the whole month →
Switch from ChatGPT to Claude with import memory feature
AI · Mar 2026 · claude.com
