nFactorial
Build distributed agents that spawn more distributed sub-agents.
nFactorial is a distributed task queue for building reliable multi-agent systems. It makes the following trivial to implement:
- Agent Reliability: Automatic retries, backoff strategies, and recovery of dropped tasks from crashed workers.
- In-flight Task Management: Cancel, steer, and monitor running tasks.
- Spawning Sub Agents: Having an agent spawn multiple sub agents and wait for their completion before continuing.
- Deferred Tools: Pause the agent while it waits for long running tools to complete externally or wait for user approval before continuing.
- Observability: Built-in metrics dashboard and comprehensive logging
Installation
pip install nfactorial
Quick Start
from factorial import Agent, Orchestrator, AgentWorkerConfig, gpt_41
def get_weather(location: str) -> str:
return f"The weather in {location} is sunny and 72°F"
agent = Agent(
instructions="You help users get weather information.",
model=gpt_41,
tools=[get_weather],
)
# Create orchestrator
orchestrator = Orchestrator(
redis_host="localhost",
redis_port=6379,
redis_db=0,
redis_max_connections=50,
)
orchestrator.register_runner(
agent=agent, agent_worker_config=AgentWorkerConfig(workers=1)
)
# Run the system
if __name__ == "__main__":
orchestrator.run()
Usage Examples:
IDE Coding Agent
/docs/examples/code_agent
Multi-Agent
/docs/examples/multi_agent
Deliberations.ai
1 maintainer
Download files
Download the file for your platform. If you're not sure which to choose, learn more about installing packages.
Source Distribution
nfactorial-0.1.30.tar.gz
(77.0 kB
view details)
Built Distribution
Filter files by name, interpreter, ABI, and platform.
If you're not sure about the file name format, learn more about wheel file names.
Copy a direct link to the current filters
File details
Details for the file nfactorial-0.1.30.tar.gz.
File metadata
- Download URL: nfactorial-0.1.30.tar.gz
- Upload date:
- Size: 77.0 kB
- Tags: Source
- Uploaded using Trusted Publishing? No
- Uploaded via:
twine/6.1.0 CPython/3.11.0
File hashes
| Algorithm | Hash digest | |
|---|---|---|
| SHA256 |
e5d94c0610ddde064443325db5afa4cdcae44371c7f6d7bbecc29d9f50348d0c
|
|
| MD5 |
5eb460aeb922b48ff390f39660bca33e
|
|
| BLAKE2b-256 |
4a09fe7156efefef8853cac6b4c546bf2c36d776e198949d312c725e1388d77c
|
File details
Details for the file nfactorial-0.1.30-py3-none-any.whl.
File metadata
- Download URL: nfactorial-0.1.30-py3-none-any.whl
- Upload date:
- Size: 88.9 kB
- Tags: Python 3
- Uploaded using Trusted Publishing? No
- Uploaded via:
twine/6.1.0 CPython/3.11.0
File hashes
| Algorithm | Hash digest | |
|---|---|---|
| SHA256 |
5e46166d5c345e4e78b9fe04dd87821e306cc9eff1fd044d84be605a1c7e971d
|
|
| MD5 |
216895a978403634b00d008270d36b8d
|
|
| BLAKE2b-256 |
12cc14b12be98252157b8ee1610d7a89b99d52cd125032a8258807851f3798a1
|
