Learn Anything about AI in one site

The `q` CLI lets developers chat with Amazon Q Developer from the command line, generating commands and code with AWS context.

Official Cloudflare MCP servers that let AI agents read, analyse and even change Cloudflare resources through natural-language requests.

Distributed KV-cache store & transfer engine that decouples prefilling from decoding to scale vLLM serving clusters.

Provides cloud-browser automation so LLMs can navigate, scrape and interact with the web via Browserbase & Stagehand.

Volcano Engine Reinforcement Learning library for efficient LLM post-training—open-sourced HybridFlow.

RAGFlow is InfiniFlow’s open-source Retrieval-Augmented Generation engine focused on deep-document understanding and scalable multi-format ingestion.

Cherry Studio is a cross-platform desktop client that unifies cloud and local LLMs, knowledge-base RAG, AI drawing and translation inside one extensible interface.

This paper introduces DeepSeek-V3, a 671B-parameter Mixture-of-Experts (MoE) language model that activates only 37B parameters per token for efficient training and inference. By leveraging innovations like Multi-head Latent Attention, auxiliary-loss-free load balancing, and multi-token prediction, it achieves top-tier performance across math, code, multilingual, and reasoning tasks. Despite its massive scale, DeepSeek-V3 maintains economical training costs and outperforms all other open-source models, achieving results comparable to leading closed-source models like GPT-4o and Claude-3.5, thereby significantly narrowing the open-source vs. closed-source performance gap.

OpenHands is an open-source platform for building full-stack AI software-development agents.

This paper introduces DeepSeek-R1, a large language model that improves reasoning purely through reinforcement learning (RL), even without supervised fine-tuning. It shows that reasoning skills like chain-of-thought, self-reflection, and verification can naturally emerge from RL, achieving performance comparable to OpenAI’s top models. Its distilled smaller models outperform many open-source alternatives, democratizing advanced reasoning for smaller systems. The work impacts the field by proving RL-alone reasoning is viable and by open-sourcing both large and distilled models, opening new directions for scalable, cost-effective LLM training and future development in reasoning-focused AI systems.

Aider is a terminal-first AI pair-programming tool that edits your local Git repo through chat.

The best introduction to how large language models (LLMs) like ChatGPT works in the world. It covers the three main stages of their training: pre-training on vast amounts of internet text, supervised fine-tuning to become helpful assistants, and reinforcement learning to improve problem-solving skills. The video also discusses LLM psychology, including why they hallucinate, how they use tools, and their limitations. Finally, it looks at future capabilities like multimodality and agent-like behavior.