> ## Documentation Index
> Fetch the complete documentation index at: https://docs.agno.com/llms.txt
> Use this file to discover all available pages before exploring further.

# Upstash Vector DB

> Install dependency: - uv pip install upstash-vector.

Install dependency: - uv pip install upstash-vector

```python theme={null}
"""
Upstash Vector DB
=================

Install dependency:
- uv pip install upstash-vector

Set OpenAI and Upstash environment variables before running.
"""

import asyncio
import os
from os import getenv

from agno.agent import Agent
from agno.knowledge.embedder.openai import OpenAIEmbedder
from agno.knowledge.knowledge import Knowledge
from agno.models.openai import OpenAIChat
from agno.vectordb.upstashdb import UpstashVectorDb


# ---------------------------------------------------------------------------
# Setup
# ---------------------------------------------------------------------------
def create_sync_knowledge() -> tuple[Knowledge, UpstashVectorDb]:
    vector_db = UpstashVectorDb(
        url=os.getenv("UPSTASH_VECTOR_REST_URL"),
        token=os.getenv("UPSTASH_VECTOR_REST_TOKEN"),
    )
    knowledge = Knowledge(
        name="Basic SDK Knowledge Base",
        description="Agno 2.0 Knowledge Implementation with Upstash Vector DB",
        vector_db=vector_db,
    )
    return knowledge, vector_db


def create_async_batch_knowledge() -> Knowledge:
    return Knowledge(
        vector_db=UpstashVectorDb(
            url=getenv("UPSTASH_VECTOR_URL", ""),
            token=getenv("UPSTASH_VECTOR_TOKEN", ""),
            dimension=1536,
            embedder=OpenAIEmbedder(enable_batch=True),
        )
    )


# ---------------------------------------------------------------------------
# Create Agent
# ---------------------------------------------------------------------------
def create_sync_agent(knowledge: Knowledge) -> Agent:
    return Agent(knowledge=knowledge)


def create_async_batch_agent(knowledge: Knowledge) -> Agent:
    return Agent(
        model=OpenAIChat(id="gpt-5.2"),
        knowledge=knowledge,
        search_knowledge=True,
        read_chat_history=True,
    )


# ---------------------------------------------------------------------------
# Run Agent
# ---------------------------------------------------------------------------
def run_sync() -> None:
    knowledge, vector_db = create_sync_knowledge()
    knowledge.insert(
        name="Recipes",
        url="https://agno-public.s3.amazonaws.com/recipes/ThaiRecipes.pdf",
        metadata={"doc_type": "recipe_book"},
    )

    agent = create_sync_agent(knowledge)
    agent.print_response("How to make Pad Thai?", markdown=True)

    vector_db.delete_by_name("Recipes")
    vector_db.delete_by_metadata({"doc_type": "recipe_book"})


async def run_async_batch() -> None:
    knowledge = create_async_batch_knowledge()
    agent = create_async_batch_agent(knowledge)

    await knowledge.ainsert(path="cookbook/07_knowledge/testing_resources/cv_1.pdf")
    await agent.aprint_response(
        "What can you tell me about the candidate and what are his skills?",
        markdown=True,
    )


if __name__ == "__main__":
    run_sync()
    asyncio.run(run_async_batch())
```

## Run the Example

```bash theme={null}
# Clone and setup repo
git clone https://github.com/agno-agi/agno.git
cd agno/cookbook/07_knowledge/vector_db/upstash_db

# Create and activate virtual environment
./scripts/demo_setup.sh
source .venvs/demo/bin/activate

# Export relevant API keys
export UPSTASH_VECTOR_REST_TOKEN="***"
export UPSTASH_VECTOR_REST_URL="***"

python upstash_db.py
```