Prompts2026年4月7日·1 分钟阅读

Agno — Lightweight AI Agent Framework for Python

Build AI agents in 5 lines of Python. Agno provides model-agnostic agents with tools, memory, knowledge bases, and team coordination at 10x less overhead.

What is Agno?

Agno (formerly Phidata) is a lightweight Python AI agent framework supporting 23+ model providers, built-in tools, knowledge bases, memory, and multi-agent teams.

In one sentence: Agno is a lightweight AI agent framework (formerly Phidata) — build production agents in 5 lines, with 23+ model providers, tools, RAG, and multi-agent teams — 20k+ GitHub stars.

For: Python developers who want a simple yet powerful agent framework. Supports: Claude, GPT, Gemini, Ollama, and 23+ other providers.

Core Features

1. Multi-Provider Support

Switch models with one line of code.

2. Built-In Tools

Search, stocks, databases, files, email, and more — out of the box.

3. Knowledge Base (RAG)

Use PDFs, web pages, and databases as knowledge sources.

4. Memory and Storage

Cross-session memory with PostgreSQL persistence.

5. Multi-Agent Teams

Multiple agents coordinate to complete complex tasks.

6. Structured Output

Define output with Pydantic models.

FAQ

Q: What happened to Phidata? A: Renamed to Agno in early 2026 — the API is largely unchanged.

Q: How does it compare to LangChain? A: Agno is lighter and more direct — faster to get started with fewer abstractions.

🙏

来源与感谢

agno-agi/agno — 20k+ stars, MPL-2.0

讨论

登录后参与讨论。
还没有评论,来写第一条吧。