Parallel

Enable Agno agents with web search and extraction infrastructure from Parallel that prioritizes token efficiency and multi-hop reasoning.

Search

Natural-language web search that returns LLM-optimized excerpts. Use when the model needs current facts, specific entities, or web data to ground a response.

from agno.agent import Agent
from agno.models.openai import OpenAIResponses
from agno.tools.parallel import ParallelTools

agent = Agent(
    model=OpenAIResponses(id="gpt-5.4"),
    tools=[ParallelTools(
        max_results=10,
        include_domains=["techcrunch.com", "wired.com"],
    )],
    markdown=True,
)

agent.print_response("What are the latest developments in AI agents?", stream=True)

Task

Deep research that takes a plain-language input and returns comprehensive, cited results. Choose the processor by complexity and latency. base typically completes in 15 to 100 seconds, while higher tiers can take minutes.

from agno.agent import Agent
from agno.models.openai import OpenAIResponses
from agno.tools.parallel import ParallelTools

enrichment_tools = ParallelTools(
    enable_search=False,
    enable_extract=False,
    enable_task=True,
    default_processor="base",
    default_output_schema={
        "type": "json",
        "json_schema": {
            "type": "object",
            "properties": {
                "company_name": {"type": "string"},
                "headquarters": {"type": "string"},
                "total_funding": {"type": "string"},
                "key_investors": {"type": "array", "items": {"type": "string"}},
            },
            "required": ["company_name"],
        },
    },
)

agent = Agent(
    model=OpenAIResponses(id="gpt-5.4"),
    tools=[enrichment_tools],
    markdown=True,
    instructions="Use create_task() to research, then get_task_result() to retrieve.",
)

agent.print_response("Research Anthropic and return structured company data", stream=True)

Monitor

Continuously track the web for changes relevant to a natural-language query, on a schedule you control. Use for news tracking, regulatory watchlists, or competitor monitoring.

from agno.agent import Agent
from agno.models.openai import OpenAIResponses
from agno.tools.parallel import ParallelTools

monitor_tools = ParallelTools(
    enable_search=False,
    enable_extract=False,
    enable_monitor=True,
    default_monitor_frequency="1d",
)

agent = Agent(
    model=OpenAIResponses(id="gpt-5.4"),
    tools=[monitor_tools],
    markdown=True,
)

# Create monitors
agent.print_response("Create a monitor to track OpenAI product launches", stream=True)

# Later: check for events
agent.print_response("List my monitors and fetch recent events", stream=True)

Run the Examples

Clone and set up

git clone --branch v2.7.4 --depth 1 https://github.com/agno-agi/agno.git
cd agno

Create virtual environment

./scripts/demo_setup.sh
source .venvs/demo/bin/activate

Export the required API keys

export PARALLEL_API_KEY="your_parallel_api_key_here"
export OPENAI_API_KEY="your_openai_api_key_here"

$Env:PARALLEL_API_KEY="your_parallel_api_key_here"
$Env:OPENAI_API_KEY="your_openai_api_key_here"

Run an example

# Search
python cookbook/91_tools/parallel/news_search.py

# Task (deep research)
python cookbook/91_tools/parallel/company_enrichment.py

# Monitor (continuous tracking)
python cookbook/91_tools/parallel/competitor_tracker.py

For more examples, see the Parallel cookbook.

Examples

Primitives

Context

Models

Tools

More

Search

Task

Monitor

Run the Examples

​Search

​Task

​Monitor

​Run the Examples

Search

Task

Monitor

Run the Examples