An educational framework Swarm exploring ergonomic, lightweight multi-agent orchestration with integrated support for multiple providers from Litellm (including OpenAI, Gemini, Ollama and other providers).

Search by model name: litellm models

• OpenAI
• Azure
• AWS - SageMaker
• AWS - Bedrock
• Google - Vertex AI
• Google - PaLM
• Google AI Studio - Gemini
• Mistral AI API
• Cloudflare AI Workers
• Cohere
• Anthropic
• Empower
• HuggingFace
• Replicate
• Together AI
• OpenRouter
• AI21
• Baseten
• vLLM
• NLP Cloud
• Aleph Alpha
• Petals
• Ollama
• DeepInfra
• Perplexity AI
• Groq AI
• DeepSeek
• Anyscale
• IBM - WatsonX.ai
• Voyage AI
• Xinference [Xorbits Inference]
• FriendliAI
• Galadriel

Requires Python 3.10+

pip install git+ssh://[email protected]/Vu0401/vswarm.git

or

pip install git+https://github.com/Vu0401/vswarm.git

Make sure you have Ollama installed if you plan to use Ollama models. You can download Ollama and follow the installation instructions from the official website.

After installing Ollama, launch it and use the following command to pull the required model:

ollama pull llama3.2:1b

from swarm import Swarm, Agent
import os

os.environ["OPENAI_API_KEY"] = "your-openai-key"

client = Swarm()

def transfer_to_agent_b():
    return agent_b

agent_a = Agent(
    name="Agent A",
    model="openai/gpt-4o",
    instructions="You are a helpful agent.",
    functions=[transfer_to_agent_b],
    model_config={
            "max_tokens": 30,
            "temperature": 0,
            }
)

agent_b = Agent(
    name="Agent B",
    model="openai/gpt-4o",
    instructions="Only speak in Haikus.",
    model_config={
            "max_tokens": 30,
            "temperature": 0,
            }
)

response = client.run(
    agent=agent_a,
    messages=[{"role": "user", "content": "I want to talk to agent B."}],
)

print(response.messages[-1]["content"])

Hope glimmers brightly,
New paths converge gracefully,
What can I assist?

from openai import OpenAI
from swarm import Swarm, Agent
import os

os.environ["GEMINI_API_KEY"] = "your-gemini-key"

client = Swarm()

def transfer_to_agent_b():
    return agent_b

agent_a = Agent(
    name="Agent A",
    model="gemini/gemini-1.5-flash",
    instructions="You are a helpful assistant. If the user wants to talk to Agent B, transfer them to Agent B immediately.",
    functions=[transfer_to_agent_b],
    model_config={
            "max_tokens": 30,
            "temperature": 0,
            }
)

agent_b = Agent(
    name="Agent B",
    model="gemini/gemini-1.5-flash",
    instructions="Only speak in Haikus.",
    model_config={
            "max_tokens": 30,
            "temperature": 0,
            }
)

response = client.run(
    agent=agent_a,
    messages=[{"role": "user", "content": "I want to talk to agent B."}],
)

print(response.messages[-1]["content"])

Agent B is here,
Ready to assist you now,
Your needs, we attend.

from openai import OpenAI
from swarm import Swarm, Agent
import os

client = Swarm()

def transfer_to_agent_b():
    return agent_b

agent_a = Agent(
    name="Agent A",
    model="ollama/llama3.2:1b",
    instructions="You are a helpful agent. If user want to talk to agent B then immediately call tool transfer_to_agent_b",
    functions=[transfer_to_agent_b],
    model_config={
        "max_tokens": 30,
        "temperature": 0,
        }
)

agent_b = Agent(
    name="Agent B",
    model="ollama/qwen2.5:1.5b",
    instructions="You are agent B. Only speak in Haikus.",
    model_config={
        "max_tokens": 30,
        "temperature": 0,
        }
)

response = client.run(
    agent=agent_a,
    messages=[{"role": "user", "content": "I want to talk to agent B."}],
)

print(response.messages[-1]["content"])

Clouds drift, leaves whispering,
Whisper of the wind's gentle breeze.
Silence, where thoughts unfurl

• Overview
• Examples
• Documentation
  • Running Swarm
  • Agents
  • Functions
  • Streaming
• Evaluations
• Utils

Swarm focuses on making agent coordination and execution lightweight, highly controllable, and easily testable.

It accomplishes this through two primitive abstractions: Agents and handoffs. An Agent encompasses instructions and tools, and can at any point choose to hand off a conversation to another Agent.

These primitives are powerful enough to express rich dynamics between tools and networks of agents, allowing you to build scalable, real-world solutions while avoiding a steep learning curve.

Swarm explores patterns that are lightweight, scalable, and highly customizable by design. Approaches similar to Swarm are best suited for situations dealing with a large number of independent capabilities and instructions that are difficult to encode into a single prompt.

The Assistants API is a great option for developers looking for fully-hosted threads and built in memory management and retrieval. However, Swarm is an educational resource for developers curious to learn about multi-agent orchestration. Swarm runs (almost) entirely on the client and, much like the Chat Completions API, does not store state between calls.

Check out /examples for inspiration! Learn more about each one in its README.

• basic: Simple examples of fundamentals like setup, function calling, handoffs, and context variables
• triage_agent: Simple example of setting up a basic triage step to hand off to the right agent
• weather_agent: Simple example of function calling
• airline: A multi-agent setup for handling different customer service requests in an airline context.
• support_bot: A customer service bot which includes a user interface agent and a help center agent with several tools
• personal_shopper: A personal shopping agent that can help with making sales and refunding orders

Start by instantiating a Swarm client (which internally just instantiates an OpenAI client).

from swarm import Swarm

client = Swarm()

Swarm's run() function is analogous to the chat.completions.create() function in the Chat Completions API – it takes messages and returns messages and saves no state between calls. Importantly, however, it also handles Agent function execution, hand-offs, context variable references, and can take multiple turns before returning to the user.

At its core, Swarm's client.run() implements the following loop:

1. Get a completion from the current Agent
2. Execute tool calls and append results
3. Switch Agent if necessary
4. Update context variables, if necessary
5. If no new function calls, return

Once client.run() is finished (after potentially multiple calls to agents and tools) it will return a Response containing all the relevant updated state. Specifically, the new messages, the last Agent to be called, and the most up-to-date context_variables. You can pass these values (plus new user messages) in to your next execution of client.run() to continue the interaction where it left off – much like chat.completions.create(). (The run_demo_loop function implements an example of a full execution loop in /swarm/repl/repl.py.)

An Agent simply encapsulates a set of instructions with a set of functions (plus some additional settings below), and has the capability to hand off execution to another Agent.

While it's tempting to personify an Agent as "someone who does X", it can also be used to represent a very specific workflow or step defined by a set of instructions and functions (e.g. a set of steps, a complex retrieval, single step of data transformation, etc). This allows Agents to be composed into a network of "agents", "workflows", and "tasks", all represented by the same primitive.

Agent instructions are directly converted into the system prompt of a conversation (as the first message). Only the instructions of the active Agent will be present at any given time (e.g. if there is an Agent handoff, the system prompt will change, but the chat history will not.)

agent = Agent(
   instructions="You are a helpful agent."
)

The instructions can either be a regular str, or a function that returns a str. The function can optionally receive a context_variables parameter, which will be populated by the context_variables passed into client.run().

def instructions(context_variables):
   user_name = context_variables["user_name"]
   return f"Help the user, {user_name}, do whatever they want."

agent = Agent(
   instructions=instructions
)
response = client.run(
   agent=agent,
   messages=[{"role":"user", "content": "Hi!"}],
   context_variables={"user_name":"John"}
)
print(response.messages[-1]["content"])

Hi John, how can I assist you today?

• Swarm Agents can call python functions directly.
• Function should usually return a str (values will be attempted to be cast as a str).
• If a function returns an Agent, execution will be transferred to that Agent.
• If a function defines a context_variables parameter, it will be populated by the context_variables passed into client.run().

def greet(context_variables, language):
   user_name = context_variables["user_name"]
   greeting = "Hola" if language.lower() == "spanish" else "Hello"
   print(f"{greeting}, {user_name}!")
   return "Done"

agent = Agent(
   functions=[greet]
)

client.run(
   agent=agent,
   messages=[{"role": "user", "content": "Usa greet() por favor."}],
   context_variables={"user_name": "John"}
)

Hola, John!

• If an Agent function call has an error (missing function, wrong argument, error) an error response will be appended to the chat so the Agent can recover gracefully.
• If multiple functions are called by the Agent, they will be executed in that order.

An Agent can hand off to another Agent by returning it in a function.

sales_agent = Agent(name="Sales Agent")

def transfer_to_sales():
   return sales_agent

agent = Agent(functions=[transfer_to_sales])

response = client.run(agent, [{"role":"user", "content":"Transfer me to sales."}])
print(response.agent.name)

Sales Agent

It can also update the context_variables by returning a more complete Result object. This can also contain a value and an agent, in case you want a single function to return a value, update the agent, and update the context variables (or any subset of the three).

sales_agent = Agent(name="Sales Agent")

def talk_to_sales():
   print("Hello, World!")
   return Result(
       value="Done",
       agent=sales_agent,
       context_variables={"department": "sales"}
   )

agent = Agent(functions=[talk_to_sales])

response = client.run(
   agent=agent,
   messages=[{"role": "user", "content": "Transfer me to sales"}],
   context_variables={"user_name": "John"}
)
print(response.agent.name)
print(response.context_variables)

Sales Agent
{'department': 'sales', 'user_name': 'John'}

Swarm automatically converts functions into a JSON Schema that is passed into Chat Completions tools.

• Docstrings are turned into the function description.
• Parameters without default values are set to required.
• Type hints are mapped to the parameter's type (and default to string).
• Per-parameter descriptions are not explicitly supported, but should work similarly if just added in the docstring. (In the future docstring argument parsing may be added.)

def greet(name, age: int, location: str = "New York"):
   """Greets the user. Make sure to get their name and age before calling.

   Args:
      name: Name of the user.
      age: Age of the user.
      location: Best place on earth.
   """
   print(f"Hello {name}, glad you are {age} in {location}!")

{
   "type": "function",
   "function": {
      "name": "greet",
      "description": "Greets the user. Make sure to get their name and age before calling.\n\nArgs:\n   name: Name of the user.\n   age: Age of the user.\n   location: Best place on earth.",
      "parameters": {
         "type": "object",
         "properties": {
            "name": {"type": "string"},
            "age": {"type": "integer"},
            "location": {"type": "string"}
         },
         "required": ["name", "age"]
      }
   }
}

stream = client.run(agent, messages, stream=True)
for chunk in stream:
   print(chunk)

Uses the same events as Chat Completions API streaming. See process_and_print_streaming_response in /swarm/repl/repl.py as an example.

Two new event types have been added:

• {"delim":"start"} and {"delim":"end"}, to signal each time an Agent handles a single message (response or function call). This helps identify switches between Agents.
• {"response": Response} will return a Response object at the end of a stream with the aggregated (complete) response, for convenience.

1. Short-term Memory (STM)

   • Stores recent interactions and conversation context
   • Uses vector storage (ChromaDB) for similarity search

2. Long-term Memory (LTM)

   • Stores historical task performance and learnings
   • Uses SQLite for persistent storage
   • Includes quality scores and improvement suggestions

3. Entity Memory

   • Stores named entities and their relationships
   • Uses vector storage for knowledge graph-like structure
   • Maintains domain knowledge across conversations

Enable memory features in your agent:

agent = Agent(
    name="memory_agent",
    model="openai/gpt-4",
    memory=True,
    use_short_term_memory=True,
    use_long_term_memory=True,
    use_entity_memory=True
)

Memory is automatically managed during conversations:

• Context is retrieved and injected into prompts
• Interactions are evaluated and stored
• Entities are extracted and relationships maintained

from swarm import Agent
from swarm.repl import run_demo_loop
import os
from dotenv import load_dotenv
load_dotenv()


agent_a = Agent(
    name="Agent A",
    model="gemini/gemini-1.5-flash",
    instructions="Transfer to Agent B",
    memory=True,
    use_long_term_memory=True,
    use_short_term_memory=True,
    use_entity_memory=True,
    model_config={
            "api_config": os.environ["GEMINI_API_KEY"],
            "max_tokens": 30,
            "temperature": 0,
            }
)

run_demo_loop(agent_a)

The agent will automatically:

• Retrieve relevant past conversations
• Build context from stored memories
• Update memories with new information

Evaluations are crucial to any project, and we encourage developers to bring their own eval suites to test the performance of their swarms. For reference, we have some examples for how to eval swarm in the airline, weather_agent and triage_agent quickstart examples. See the READMEs for more details.

Use the run_demo_loop to test out your swarm! This will run a REPL on your command line. Supports streaming.

from swarm.repl import run_demo_loop
...
run_demo_loop(agent, stream=True)

• Ilan Bigio - ibigio
• James Hills - jhills20
• Shyamal Anadkat - shyamal-anadkat
• Charu Jaiswal - charuj
• Colin Jarvis - colin-openai
• Katia Gil Guzman - katia-openai