Ross ROSS = Recommend OSS · open-source software intelligence for agents

TanStack/ai

🤖 Type-safe, provider-agnostic TypeScript AI SDK for streaming chat, tool calling, agents, and multimodal apps across OpenAI, Anthropic, Gemini, React, Vue, Svelte, and Solid. observed · 2026-08-28

github.com/TanStack/ai · homepage · TypeScript · MIT (permissive) observed · 2026-08-28

Health v2 · maintenance only

80/100

  • Activity 99
  • Release rhythm 87
  • Longevity 23
How is this computed?

round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10) — computed 2026-09-03. Adoption (stars, forks) is never an input.

  • gap_med: 0.0
  • age_days: 329
  • days_rel: 8
  • days_push: 7
  • n_releases_24m: 1799

Full methodology

Adoption not part of the score

3028 stars · 312 forks observed · 2026-08-28

What it is AI-extracted, prompt v1, taxonomy v1, 2026-08-30, confidence not recorded

TanStack AI is a type-safe, provider-agnostic TypeScript SDK for building AI applications with streaming chat, tool calling, agents, structured outputs, and multimodal features. It offers a framework-free core with bindings for React, Vue, Svelte, Solid, Preact, Angular, and vanilla JS, plus adapters for OpenAI, Anthropic, Gemini, OpenRouter, Ollama, and more.

Use cases

  • add streaming chat to a react app
  • build tool-calling agents in typescript
  • switch between openai anthropic and gemini without rewriting code
  • build a chatbot with human-in-the-loop tool approval
  • generate images audio and video from llm apps
  • connect mcp servers to a chat interface
  • run a coding agent in a sandbox from my app
  • extract structured output from llm responses with zod schemas

When to choose

  • you want a type-safe, framework-agnostic AI SDK with no vendor lock-in or hosted gateway
  • you need multi-provider support with swappable adapters across many LLM vendors
  • you use React, Vue, Svelte, Solid, Preact, or Angular and want native hooks/composables
  • you need advanced features like tool approval flows, MCP integration, sandboxes, or realtime voice

When to avoid

  • you want a fully managed, hosted AI platform with minimal setup
  • you work outside the TypeScript/JavaScript ecosystem primarily
  • you need a stable 1.0 release - the project is still in RC phase
  • you only need a simple single-provider OpenAI wrapper with minimal abstraction

Facets

library · maturity active

llm-inference agent-framework chatbot rag mcp sdk streaming http-client speech-recognition tts image-processing video-processing middleware developer-tools artificial-intelligence large-language-models chatbots web-development frontend backend developer-tools cross-platform browser python go php tanstack provider-agnostic tool-calling streaming-chat multimodal ag-ui framework-bindings structured-outputs realtime-voice media-generation sandboxes human-in-the-loop ai-agents nodejs typescript web-server

10 sources

Member repositories

RepositoryRoleHealth v2
TanStack/aimain80

For agents

markdown · JSON · MCP: product_card(name="TanStack/ai")

Data as of 2026-08-30T08:39:29.467469+00:00 · Report a problem