AI

Learn Claude Code offers structured tutorials and code examples to help developers understand and build AI agent harnesses from scratch. It covers engineering practices from basic agent loops to multi-agent systems, with Python implementations and a web learning platform.

Unified LLM API gateway giving developers one OpenAI-compatible endpoint to 400+ models from 70+ providers, with auto-routing, fallbacks and pay-as-you-go pricing to cut lock-in and improve uptime.

LM Studio is a desktop application that lets you discover, download, and run open-source large language models on your local computer. It offers a chat interface, OpenAI-compatible API, and developer tools for privacy-focused AI experimentation and development.

Vibe Vibe is an open-source tutorial platform teaching AI-powered creation (Vibe Coding) for everyone. It guides beginners with no coding experience to turn ideas into working prototypes and full products using natural language and AI tools.

AITMPL is a comprehensive template library for Claude Code featuring hundreds of pre-built AI agents, slash commands, skills, MCP integrations, hooks, and settings. Install via CLI to quickly enhance your Anthropic Claude Code development workflow.

Arena AI is a community-powered platform for evaluating and ranking AI models. Users submit prompts in anonymous Battle mode to compare LLMs, image, and multimodal models side-by-side, voting to build transparent, real-world leaderboards.

bbycroft.net/llm is an interactive 3D visualization tool that demonstrates the inner workings of GPT-style large language models using a small nano-gpt example. It helps learners understand token embeddings, attention mechanisms, transformer layers, and more through animated walkthroughs.

Ollama is an open-source tool that makes it easy to run large language models locally on your computer. It offers full data privacy, offline capabilities, and optional cloud model acceleration. Ideal for developers, researchers, and privacy-conscious users building AI applications.

Midjourney is an AI image and short video generation tool for creators, designers, and content teams who need to explore visual concepts, styles, and creative drafts from prompts or reference images.

DeepSeek offers an AI chat assistant and developer API for writing, reasoning, coding, and model integration, serving both everyday users and technical teams.

A community platform for data scientists and ML practitioners offering competitions, public datasets, cloud notebooks with free GPU/TPU, and free courses to practice modeling and improve skills.

Chris Olah’s personal technical blog offering clear visual explanations of neural networks, LSTMs, convolutions, and interpretability for researchers and learners.

Open-source high-throughput LLM inference and serving engine. Uses PagedAttention for efficient KV cache management, continuous batching, and an OpenAI-compatible API for deploying open models.

Open-source minimal terminal AI coding agent harness. Ships four core tools and supports 15+ model providers. Extend with TypeScript extensions, skills, and packages to fit your own workflow.

Injects up-to-date, version-specific library docs and code examples into Cursor, Claude and other AI coding agents to reduce hallucinations and outdated APIs.

NVIDIA’s developer platform for free hosted NIM inference APIs, optimized models, and AI Blueprints—ideal for prototyping LLM apps and agentic workflows without GPU setup.

fast.ai offers free practical deep learning courses, the fastai library for PyTorch, and supporting resources to help coders quickly learn and apply deep learning to real-world problems.

Independent benchmarking platform for AI models and API providers. Compare models across intelligence, speed, cost, and other metrics to help developers, researchers, and businesses choose the best AI technology for their use cases.

Scale Labs is Scale AI's research platform offering benchmarks, leaderboards, papers, and insights on frontier AI models. It focuses on agents, evaluation, safety, alignment, and real-world performance to help developers, researchers, and enterprises track AI advancements.

MiniMax is a leading Chinese AI company offering frontier multimodal models including M3 for coding and agents, Hailuo video generation, speech and music models, and an open API platform. Features long context and strong agent capabilities for developers and creators.

Z.ai is the international platform for the GLM series of large language models, offering a chat interface and API. Supports coding, agentic tasks, and multimodal capabilities for developers building applications and automated workflows.