Memory for AI Agents in 6 lines of code

Intelligent Mixture-of-Models Router for Efficient LLM Inference

Optimize prompts, code, and more with AI-powered Reflective Text Evolution

Build an email assistant with human-in-the-loop and memory

📰 Must-read papers and blogs on LLM based Long Context Modeling 🔥

Environments for LLM Reinforcement Learning

Distributed LLM inference. Connect home devices into a powerful cluster to accelerate LLM inference. More devices means faster inference.

Learning Deep Representations of Data Distributions

Achieve state of the art inference performance with modern accelerators on Kubernetes

Multilingual Document Layout Parsing in a Single Vision-Language Model

An open-source tool for LLM prompt optimization.

Build Real-Time Knowledge Graphs for AI Agents

A k6 extension for Server-Sent Events (SSE)

🔥 Reliable Browser AI Agents (YC S25)

Renderer for the harmony response format to be used with gpt-oss

All information and news with respect to Falcon-H1 series

SSH to any machine without ip

WebGL globe for displaying websocket JSON realtime data and static markers on the Earth's surface

Main community repo for agentic-community documents and discussions. Onboarding guide:

Hierarchical Reasoning Model Official Release

Kimi K2 is the large language model series developed by Moonshot AI team

Simple & Scalable Pretraining for Neural Architecture Research

vLLM’s reference system for K8S-native cluster-wide deployment with community-driven performance optimization

Supercharge Your LLM with the Fastest KV Cache Layer

Prompts for our Grok chat assistant and the `@grok` bot on X.

Implementation of the Google Zero-Knowledge library for Identity Protocols.

A program to display status from the Enphase Combiner