mirror of
https://github.com/qdrant/landing_page.git
synced 2026-10-04 10:28:29 +02:00
72668d521e
* docs: add agno framework integration documentation * Update qdrant-landing/content/documentation/frameworks/agno.md Co-authored-by: Copilot <175728472+Copilot@users.noreply.github.com> * Apply suggestion from @Anush008 Co-authored-by: Anush <anushshetty90@gmail.com> * docs: remove inline-example & add agno to frameworks list --------- Co-authored-by: Copilot <175728472+Copilot@users.noreply.github.com> Co-authored-by: Anush <anushshetty90@gmail.com>
73 lines
2.1 KiB
Markdown
73 lines
2.1 KiB
Markdown
---
|
|
title: Agno
|
|
---
|
|
|
|
# Agno
|
|
|
|
[Agno](https://github.com/agno-agi/agno) is an incredibly fast multi-agent framework, runtime and UI. It enables you to build multi-agent systems with memory, knowledge, human-in-the-loop capabilities, and Model Context Protocol (MCP) support.
|
|
|
|
You can orchestrate agents as multi-agent teams (providing more autonomy) or step-based agentic workflows (offering more control). Agno works seamlessly with Qdrant as a vector database for knowledge bases, enabling efficient storage and retrieval of information for your AI agents.
|
|
|
|
Agno supports both synchronous and asynchronous operations, making it flexible for various use cases and deployment scenarios.
|
|
|
|
## Usage
|
|
|
|
- Install the required dependencies
|
|
|
|
```bash
|
|
pip install agno qdrant-client
|
|
```
|
|
|
|
- Set up environment variables for Qdrant connection
|
|
|
|
```bash
|
|
export QDRANT_API_KEY="<your-qdrant-api-key>"
|
|
export QDRANT_URL="<your-qdrant-url>"
|
|
```
|
|
|
|
- Create an agent with Qdrant knowledge base
|
|
|
|
```python
|
|
import os
|
|
from agno.agent import Agent
|
|
from agno.knowledge.knowledge import Knowledge
|
|
from agno.vectordb.qdrant import Qdrant
|
|
|
|
# Configure Qdrant vector database
|
|
api_key = os.getenv("QDRANT_API_KEY")
|
|
qdrant_url = os.getenv("QDRANT_URL")
|
|
COLLECTION_NAME = "my-knowledge-base"
|
|
|
|
vector_db = Qdrant(
|
|
collection=COLLECTION_NAME,
|
|
url=qdrant_url,
|
|
# or you can just url="http://localhost:6333"
|
|
api_key=api_key, # (optional)
|
|
)
|
|
|
|
# Create a knowledge base with Qdrant
|
|
knowledge_base = Knowledge(
|
|
vector_db=vector_db,
|
|
)
|
|
|
|
# Add content to the knowledge base
|
|
knowledge_base.add_content(
|
|
url="https://example.com/document.pdf"
|
|
)
|
|
|
|
# Create an agent with the knowledge base
|
|
agent = Agent(
|
|
knowledge=knowledge_base,
|
|
debug_mode=True,
|
|
)
|
|
|
|
# Use the agent
|
|
response = agent.print_response("What information do you have?")
|
|
```
|
|
|
|
## Further Reading
|
|
|
|
- [Agno Documentation](https://docs.agno.com/introduction)
|
|
- [Qdrant integration with Agno](https://docs.agno.com/integrations/vectordb/qdrant/overview)
|
|
- [Qdrant Asynchronous](https://docs.agno.com/integrations/vectordb/qdrant/usage/async-qdrant-db)
|
|
- [Agno GitHub Repository](https://github.com/agno-agi/agno) |