TurboQuant: Near-optimal KV cache quantization for LLM inference (3-bit keys, 2-bit values) with Triton kernels + vLLM integration

Fluent migrations framework for .NET

Zig implementation of generic parser combinators

Papers from the computer science community to read and discuss.

Deep research agents using MiniMax M2.1 interleaved thinking

a small lib to play with python

Minimal, clean code for the Byte Pair Encoding (BPE) algorithm commonly used in LLM tokenization.

Indirect and direct asynchronous programming style applied to Java 21+ and an effective actor system.

Unified Schema-Based Information Extraction

Burn is a next generation tensor library and Deep Learning Framework that doesn't compromise on flexibility, efficiency and portability.

A cross-platform, safe, pure-Rust graphics API.

IVM (Incremental View Maintenance) implementation as a PostgreSQL extension

SigNoz is an open-source observability platform native to OpenTelemetry with logs, traces and metrics in a single application. An open-source alternative to DataDog, NewRelic, etc. 🔥 🖥. 👉 Open sour…

Library for generalised parser combinators and a dedicated meta-language in Rust

[EMNLP2025] "LightRAG: Simple and Fast Retrieval-Augmented Generation"

Low latency web data collector

12 Lessons to Get Started Building AI Agents

📊 Cube Core is open-source semantic layer for AI, BI and embedded analytics

Toolkit for linearizing PDFs for LLM datasets/training

Rust multiprovider generative AI client (Ollama, OpenAi, Anthropic, Gemini, DeepSeek, xAI/Grok, Groq,Cohere, ...)

A programming framework for agentic AI

OpenAI API client library for Rust (unofficial)

NVIDIA Cosmos is an open platform of world models, datasets, and tools that enables developers to build Physical AI for robots, autonomous vehicles, smart infrastructure, and more.

HTML parsing and querying with CSS selectors