The AI stack, by layer

47 of 47 platforms

Platforms

• Agent frameworksOpen source

  AI SDK

  Vercel

  TypeScript toolkit for streaming text, structured output, and tool calls across many providers.

  typescript · streaminggithub.com
• Coding agentsOpen source

  Aider

  Open-source pair programming in the terminal that edits files and commits changes to git.

  terminal · gitaider.chat
• Model APIs

  Amazon Bedrock

  AWS

  Managed access to models from several labs inside your AWS account, with IAM, VPC, and guardrails.

  cloud · enterpriseaws.amazon.com
• Inference platforms

  Baseten

  Dedicated and shared inference for open-weight and custom models, plus training and the Truss packaging tool.

  dedicated inference · trainingbaseten.co
• GatewaysOpen source

  Bifrost

  Maxim AI

  Open-source AI gateway written in Go, with one API for many providers plus failover, caching, and governance.

  go · mcp gatewaygithub.com
• Vector databasesOpen source

  Chroma

  Embedding database that runs in-process for prototypes and as a server or cloud service in production.

  python · embeddedgithub.com
• Model APIs

  Claude API

  Anthropic

  Claude models through the Messages API, with tool use, prompt caching, and batch processing.

  frontier · prompt cachingdocs.anthropic.com
• Coding agents

  Claude Code

  Anthropic

  Agentic coding tool that works in your terminal, IDE, desktop app, and browser, and can run commands and edit files.

  terminal · agentdocs.anthropic.com
• Coding agentsOpen source

  Cline

  Open-source coding agent for VS Code that asks for approval before it edits files or runs commands.

  editor · mcpcline.bot
• Gateways

  Cloudflare AI Gateway

  Cloudflare

  Managed gateway on Cloudflare's network with caching, rate limiting, retries, and analytics.

  managed · edgedevelopers.cloudflare.com
• Coding agents

  Codex

  OpenAI

  OpenAI's coding agent, available as an open-source CLI, IDE extension, and cloud agent.

  terminal · cloud agentdevelopers.openai.com
• Agent frameworksOpen source

  CrewAI

  Python framework for teams of role-based agents and event-driven flows.

  python · multi-agentgithub.com
• Coding agents

  Cursor

  Anysphere

  AI code editor built on VS Code, with tab completion, an agent mode, and background agents.

  editor · agentcursor.com
• Model APIs

  DeepSeek API

  DeepSeek

  OpenAI-compatible API for DeepSeek's chat and reasoning models, which are also published as open weights.

  open weights · openai compatibleapi-docs.deepseek.com
• Inference platforms

  Fireworks AI

  Fast inference for open-weight models, with fine-tuning and on-demand deployments.

  open weights · fine-tuningfireworks.ai
• Model APIs

  Gemini API

  Google

  Gemini models for text, image, audio, and video input, available through Google AI Studio and Vertex AI.

  frontier · multimodalai.google.dev
• Coding agentsOpen source

  Gemini CLI

  Google

  Open-source terminal agent that runs on Gemini models and supports MCP servers.

  terminal · mcpgithub.com
• Coding agents

  GitHub Copilot

  GitHub

  Completions, chat, and a coding agent that works from GitHub issues and opens pull requests.

  editor · githubgithub.com
• Inference platforms

  Groq

  Inference API running open-weight models on Groq's own LPU chips, built for low latency.

  custom silicon · low latencygroq.com
• Inference platforms

  Hugging Face

  The main hub for open-weight models and datasets, plus Inference Providers and dedicated Inference Endpoints.

  open weights · model hubhuggingface.co
• Agent frameworksOpen source

  LangChain

  Integrations and building blocks for LLM apps: models, retrievers, tools, and agent loops.

  python · typescriptgithub.com
• Observability and evalsOpen source

  Langfuse

  Open-source LLM engineering platform for tracing, prompt management, and evaluations.

  tracing · self-hostablegithub.com
• Agent frameworksOpen source

  LangGraph

  LangChain

  Build agents as stateful graphs with checkpoints, human-in-the-loop steps, and streaming.

  python · typescriptgithub.com
• GatewaysOpen source

  LiteLLM

  BerriAI

  Python SDK and proxy server that call many providers through the OpenAI format.

  python · proxygithub.com
• Inference enginesOpen source

  llama.cpp

  ggml-org

  C/C++ inference for GGUF models on CPUs, Apple silicon, and consumer GPUs, with heavy quantization support.

  local · ggufgithub.com
• Agent frameworksOpen source

  LlamaIndex

  Framework for agents over your data, with document parsing, indexing, and retrieval built in.

  rag · pythongithub.com
• Observability and evals

  Maxim AI

  Platform for simulating, evaluating, and observing AI agents before and after release.

  evals · simulationgetmaxim.ai
• Vector databasesOpen source

  Milvus

  Zilliz

  Distributed vector database built for billions of vectors, with GPU-accelerated indexes.

  distributed · scalegithub.com
• Model APIs

  Mistral AI

  Hosted API for Mistral's models, several of which are also released as open weights.

  open weights · europedocs.mistral.ai
• Inference platforms

  Modal

  Modal Labs

  Serverless GPUs driven from Python code, used for inference, batch jobs, and fine-tuning.

  serverless · pythonmodal.com
• Inference enginesOpen source

  Ollama

  Pull and run open-weight models locally with one command, behind a local REST API.

  local · developer toolollama.com
• Agent frameworksOpen source

  OpenAI Agents SDK

  OpenAI

  Lightweight SDK for agents with handoffs, guardrails, sessions, and built-in tracing.

  python · typescriptgithub.com
• Model APIs

  OpenAI API

  OpenAI

  GPT and reasoning models with the Responses API, tool calling, structured outputs, and realtime audio.

  frontier · tool callingplatform.openai.com
• Inference platforms

  OpenRouter

  One OpenAI-compatible API and one bill for hundreds of models across many providers.

  router · openai compatibleopenrouter.ai
• Vector databasesOpen source

  pgvector

  Postgres extension for vector similarity search, so embeddings live next to your relational data.

  postgresgithub.com
• Observability and evals

  Phoenix

  Arize AI

  Source-available tracing and evaluation tool built on OpenTelemetry, which you can run locally.

  opentelemetry · tracinggithub.com
• Vector databases

  Pinecone

  Managed serverless vector database with namespaces, metadata filtering, and hybrid search.

  managed · serverlesspinecone.io
• Agent frameworksOpen source

  Pydantic AI

  Pydantic

  Type-safe Python agent framework from the Pydantic team, with validated structured outputs.

  python · typedgithub.com
• Vector databasesOpen source

  Qdrant

  Vector database written in Rust, with payload filtering and a managed cloud option.

  rust · filteringgithub.com
• Inference platforms

  Replicate

  Run community and official models through an API, and package your own with the open-source Cog tool.

  image · videoreplicate.com
• Inference enginesOpen source

  SGLang

  SGLang project

  Serving engine with RadixAttention prefix caching and a fast runtime for structured and multi-call workloads.

  prefix caching · gpu servinggithub.com
• Inference enginesOpen source

  TensorRT-LLM

  NVIDIA

  NVIDIA's library for optimized LLM inference on NVIDIA GPUs, with in-flight batching and quantization.

  nvidia · gpu servinggithub.com
• Inference platforms

  Together AI

  Serverless and dedicated inference for open-weight models, with fine-tuning and GPU clusters.

  open weights · fine-tuningtogether.ai
• Model APIs

  Vertex AI

  Google Cloud

  Google Cloud's platform for Gemini and third-party models, with tuning, evaluation, and agent tooling.

  cloud · enterprisecloud.google.com
• Inference enginesOpen source

  vLLM

  vLLM project

  High-throughput serving engine built around PagedAttention, continuous batching, and an OpenAI-compatible server.

  paged attention · gpu servinggithub.com
• Vector databasesOpen source

  Weaviate

  Vector database with built-in hybrid search and modules for generating embeddings.

  hybrid searchgithub.com
• Model APIs

  xAI API

  xAI

  Grok models through an API that is compatible with the OpenAI and Anthropic SDKs.

  frontierdocs.x.ai