AI & LLM SDKs Packages
SDKs and clients for large-language-model providers and inference, covering chat completions, embeddings, and streaming (OpenAI, Anthropic, Vercel AI).
Packages in AI & LLM SDKs
Kind
Sort:Used by Most Apps
LangChain.js Community
Third-party integrations for LangChain.js — chat models, vector stores, embeddings, tools, and document loaders for dozens of providers.
Mastra Core
The TypeScript framework for building AI agents, workflows, and RAG apps with unified access to 40+ LLM providers.
google-cloud-aiplatform
Google's official Python SDK for Vertex AI model training, deployment, and generative AI workflows
ollama-ai-provider-v2
A Vercel AI SDK provider for running local LLMs through Ollama, with streaming, tool calls, thinking mode, and embeddings.
assistant-ui
Composable React primitives and a headless runtime for building production-grade AI chat interfaces.
go-openai
An unofficial Go client for the OpenAI API, covering Responses, Chat Completions, embeddings, images, audio, files, fine-tuning, batches, and vector stores.
Tavily Python
Official Python SDK for Tavily's search, extract, crawl, and research API
anthropic-sdk-go
Official Go SDK for Claude's Messages, Files, Skills, and Beta agent APIs, with built-in Bedrock and Vertex AI support.
exa-js
Official TypeScript/JavaScript SDK for the Exa AI-native web search API
llama-index-llms-openai-like
LlamaIndex LLM integration for any OpenAI-compatible API endpoint, via the OpenAILike class.
mcp-go
The Go SDK for building Model Context Protocol servers and clients with minimal boilerplate.
OpenTelemetry Instrumentation for Google Generative AI
OpenTelemetry auto-instrumentation for Google Gemini and Generative AI calls in Python
@openai/agents
OpenAI's official multi-agent framework for JavaScript and TypeScript, with tools, handoffs, guardrails, and realtime voice.
claude-agent-acp
Use the Claude Agent SDK as a coding agent from any ACP-compatible editor
Replicate
Official JavaScript/TypeScript client for running and managing machine learning models on Replicate's cloud API.
langchain-google-genai
LangChain's official Python integration for Google's Gemini chat, embedding, and multimodal models.
hf-hub
Rust client for the Hugging Face Hub API with async and blocking interfaces
Mem0
Universal memory layer that gives AI agents and assistants long-term, personalized recall
LangSmith Python SDK
The official Python client for tracing, evaluating, and monitoring LLM applications on the LangSmith platform.
tiktoken-go
A Go port of OpenAI's tiktoken, for counting and encoding tokens against GPT models.
Cohere Python SDK
Official Python client for the Cohere API, with typed access to chat, embed, rerank, and generate across AWS, Azure, GCP, and OCI.
instructor
Get reliable, validated structured JSON from any LLM provider using Pydantic models and automatic retries.
Arize Phoenix
Open-source AI observability and evaluation platform for LLM apps and agents
Strands Agents
A model-driven Python SDK for building and running AI agents in just a few lines of code
