AI & Machine Learning

Official model provider SDKs plus unified clients and gateways for calling many LLMs through one interface.

OpenAI Python SDK, OpenAI Node SDK, Anthropic Python SDK

Hosted platforms for running proprietary and open-weight models via API, from hyperscalers to fast GPU clouds.

Amazon Bedrock, Microsoft Foundry, Google Vertex AI

Libraries for composing prompts, retrieval, tools and structured outputs into LLM-powered applications.

LangChain, LangChain.js, LlamaIndex

Frameworks and SDKs for building tool-using, multi-step and multi-agent LLM systems.

LangGraph, OpenAI Agents SDK, Claude Agent SDK

The Model Context Protocol, its SDKs and key servers, plus emerging standards for agent interoperability.

Model Context Protocol, MCP TypeScript SDK, MCP Python SDK

Terminal agents, AI-native IDEs and editor extensions that write, refactor, review and run code.

Claude Code, GitHub Copilot, Cursor

Run open-weight models on your own laptop, workstation or home server.

Ollama, llama.cpp, LM Studio

High-throughput engines and orchestration layers for serving models in production on GPU clusters.

vLLM, SGLang, TensorRT-LLM

Self-hostable chat interfaces and always-on personal agents that connect to local or hosted models.

Open WebUI, OpenClaw, Hermes Agent

Core libraries for training neural networks and classical machine learning models.

PyTorch, scikit-learn, JAX

The Hub and libraries most open-model work is built on: models, datasets, fine-tuning and embeddings.

Transformers, Hugging Face Hub, Datasets

Tools for fine-tuning, reinforcement learning and large-scale distributed training of models.

Unsloth, LlamaFactory, TRL

Stores and indexes for embeddings that power semantic search, RAG and recommendations.

Pinecone, pgvector, Qdrant

Turn PDFs, office files and websites into clean, LLM-ready data and retrieval pipelines.

Docling, MarkItDown, Unstructured

Infrastructure agents need beyond the LLM: long-term memory, browser control, code sandboxes and tool integrations.

Mem0, Letta, Graphiti

Tracing, prompt management, evaluation and red-teaming for LLM and agent applications.

Langfuse, LangSmith, Arize Phoenix

Track experiments, version data and models, run ML pipelines and deploy models reproducibly.

MLflow, Weights & Biases, Optuna

Visual builders for AI workflows and agents, plus Python and React toolkits for shipping AI app UIs fast.

n8n, Dify, Langflow

Open models, UIs and pipelines for generating and editing images and video.

ComfyUI, Diffusers, FLUX

Speech recognition, text-to-speech and real-time voice agent frameworks.

Whisper, faster-whisper, WhisperX

Libraries and models for image processing, detection, segmentation and visual understanding.

OpenCV, Ultralytics YOLO, torchvision