-
Notifications
You must be signed in to change notification settings - Fork 4k
Python: Multi-agent orchestration: Magentic #12104
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Merged
TaoChenOSU
merged 11 commits into
main
from
taochen/python-multi-agent-orchestration-pr-part-4
May 16, 2025
Merged
Changes from 8 commits
Commits
Show all changes
11 commits
Select commit
Hold shift + click to select a range
fd24140
Magentic orchestration
TaoChenOSU ca83971
Refine magentic manager base
TaoChenOSU 7c01aa9
Tune magentic sample
TaoChenOSU b01eb4a
Merge branch 'main' into local-branch-python-multi-agent-orchestratio…
TaoChenOSU b508bb6
Update sample readme
TaoChenOSU f0cdd3b
Tune magentic sample
TaoChenOSU 954e46c
Address comments
TaoChenOSU 1f9ed45
Merge branch 'main' into taochen/python-multi-agent-orchestration-pr-…
moonbox3 b5d52b6
Address comments 2
TaoChenOSU 122f55a
Fix mypy
TaoChenOSU 577101e
Update agent description
TaoChenOSU File filter
Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
There are no files selected for viewing
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
48 changes: 48 additions & 0 deletions
48
python/samples/getting_started_with_agents/multi_agent_orchestration/README.md
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| Original file line number | Diff line number | Diff line change |
|---|---|---|
| @@ -0,0 +1,48 @@ | ||
| # Multi-agent orchestration | ||
TaoChenOSU marked this conversation as resolved.
Show resolved
Hide resolved
|
||
|
|
||
| The Semantic Kernel Agent Framework now supports orchestrating multiple agents to work together to complete a task. | ||
|
|
||
| ## Background | ||
|
|
||
| The following samples are beneficial if you are just getting started with Semantic Kernel. | ||
|
|
||
| - [Chat Completion](../../concepts/chat_completion/) | ||
| - [Auto Function Calling](../../concepts/auto_function_calling/) | ||
| - [Structured Output](../../concepts/structured_output/) | ||
| - [Getting Started with Agents](../../getting_started_with_agents/) | ||
| - [More advanced agent samples](../../concepts/agents/) | ||
|
|
||
| ## Prerequisites | ||
|
|
||
| The following environment variables are required to run the samples: | ||
|
|
||
| - OPENAI_API_KEY | ||
| - OPENAI_CHAT_MODEL_ID | ||
|
|
||
| However, if you are using other model services, feel free to switch to those in the samples. | ||
| Refer to [here](../../concepts/setup/README.md) on how to set up the environment variables for your model service. | ||
|
|
||
| ## Orchestrations | ||
|
|
||
| | **Orchestrations** | **Description** | | ||
| | ------------------ | --------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- | | ||
| | **Concurrent** | Useful for tasks that will benefit from independent analysis from multiple agents. | | ||
| | **Sequential** | Useful for tasks that require a well-defined step-by-step approach. | | ||
| | **Handoff** | Useful for tasks that are dynamic in nature and don't have a well-defined step-by-step approach. | | ||
| | **GroupChat** | Useful for tasks that will benefit from inputs from multiple agents and a highly configurable conversation flow. | | ||
| | **Magentic** | GroupChat like with a planner based manager. Inspired by [Magentic One](https://www.microsoft.com/en-us/research/articles/magentic-one-a-generalist-multi-agent-system-for-solving-complex-tasks/). | | ||
|
|
||
| ## Samples | ||
|
|
||
| | Sample | Description | | ||
| |-----------------------------------------------------------------------------|--------------| | ||
| | [step1_concurrent](step1_concurrent.py) | Run agents in parallel on the same task. | | ||
| | [step1a_concurrent_structure_output](step1a_concurrent_structure_output.py) | Run agents in parallel on the same task and return structured output. | | ||
| | [step2_sequential](step2_sequential.py) | Run agents in sequence to complete a task. | | ||
| | [step2a_sequential_cancellation_token](step2a_sequential_cancellation_token.py) | Cancel an invocation while it is in progress. | | ||
| | [step3_group_chat](step3_group_chat.py) | Run agents in a group chat to complete a task. | | ||
| | [step3a_group_chat_human_in_the_loop](step3a_group_chat_human_in_the_loop.py) | Run agents in a group chat with human in the loop. | | ||
| | [step3b_group_chat_with_chat_completion_manager](step3b_group_chat_with_chat_completion_manager.py) | Run agents in a group chat with a more dynamic manager. | | ||
| | [step4_handoff](step4_handoff.py) | Run agents in a handoff orchestration to complete a task. | | ||
| | [step4a_handoff_structure_input](step4a_handoff_structure_input.py) | Run agents in a handoff orchestration to complete a task with structured input. | | ||
| | [step5_magentic](step5_magentic.py) | Run agents in a Magentic orchestration to complete a task. | | ||
File renamed without changes.
198 changes: 198 additions & 0 deletions
198
python/samples/getting_started_with_agents/multi_agent_orchestration/step5_magentic.py
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| Original file line number | Diff line number | Diff line change |
|---|---|---|
| @@ -0,0 +1,198 @@ | ||
| # Copyright (c) Microsoft. All rights reserved. | ||
|
|
||
| import asyncio | ||
|
|
||
| from semantic_kernel.agents import ( | ||
| Agent, | ||
| ChatCompletionAgent, | ||
| MagenticOrchestration, | ||
| OpenAIAssistantAgent, | ||
| StandardMagenticManager, | ||
| ) | ||
| from semantic_kernel.agents.runtime import InProcessRuntime | ||
| from semantic_kernel.connectors.ai.open_ai import OpenAIChatCompletion | ||
| from semantic_kernel.connectors.ai.open_ai.prompt_execution_settings.open_ai_prompt_execution_settings import ( | ||
| OpenAIChatPromptExecutionSettings, | ||
| ) | ||
| from semantic_kernel.contents import ChatMessageContent | ||
|
|
||
| """ | ||
| The following sample demonstrates how to create a Magentic orchestration with two agents: | ||
| - A Research agent that can perform web searches | ||
| - A Coder agent that can run code using the code interpreter | ||
|
|
||
| Read more about Magentic here: | ||
| https://www.microsoft.com/en-us/research/articles/magentic-one-a-generalist-multi-agent-system-for-solving-complex-tasks/ | ||
|
|
||
| This sample demonstrates the basic steps of creating and starting a runtime, creating | ||
| a Magentic orchestration with two agents and a Magentic manager, invoking the | ||
| orchestration, and finally waiting for the results. | ||
|
|
||
| The Magentic manager requires a chat completion model that supports structured output. | ||
| """ | ||
|
|
||
|
|
||
| async def agents() -> list[Agent]: | ||
| """Return a list of agents that will participate in the Magentic orchestration. | ||
|
|
||
| Feel free to add or remove agents. | ||
| """ | ||
| research_agent = ChatCompletionAgent( | ||
| name="ResearchAgent", | ||
| description="A helpful assistant with access to web search. Ask it to perform web searches.", | ||
| instructions=( | ||
| "You are a Researcher. You find information without additional computation or quantitative analysis." | ||
| ), | ||
| # This agent requires the gpt-4o-search-preview model to perform web searches. | ||
| service=OpenAIChatCompletion(ai_model_id="gpt-4o-search-preview"), | ||
TaoChenOSU marked this conversation as resolved.
Show resolved
Hide resolved
|
||
| ) | ||
|
|
||
| # Create an OpenAI Assistant agent with code interpreter capability | ||
| client, model = OpenAIAssistantAgent.setup_resources() | ||
| code_interpreter_tool, code_interpreter_tool_resources = OpenAIAssistantAgent.configure_code_interpreter_tool() | ||
| definition = await client.beta.assistants.create( | ||
| model=model, | ||
| name="CoderAgent", | ||
| description="A helpful assistant with code interpreter capability.", | ||
TaoChenOSU marked this conversation as resolved.
Show resolved
Hide resolved
|
||
| instructions="You solve questions using code. Please provide detailed analysis and computation process.", | ||
| tools=code_interpreter_tool, | ||
| tool_resources=code_interpreter_tool_resources, | ||
| ) | ||
| coder_agent = OpenAIAssistantAgent( | ||
| client=client, | ||
| definition=definition, | ||
| ) | ||
|
|
||
| return [research_agent, coder_agent] | ||
|
|
||
|
|
||
| def agent_response_callback(message: ChatMessageContent) -> None: | ||
| """Observer function to print the messages from the agents.""" | ||
| print(f"**{message.name}**\n{message.content}") | ||
|
|
||
|
|
||
| async def main(): | ||
| """Main function to run the agents.""" | ||
| # 1. Create a Magentic orchestration with two agents and a Magentic manager | ||
| # Note, the Standard Magentic manager accepts custom prompts for advanced users and scenarios. | ||
TaoChenOSU marked this conversation as resolved.
Show resolved
Hide resolved
|
||
| magentic_orchestration = MagenticOrchestration( | ||
| members=await agents(), | ||
| manager=StandardMagenticManager( | ||
| chat_completion_service=OpenAIChatCompletion(), | ||
| prompt_execution_settings=OpenAIChatPromptExecutionSettings(), | ||
| ), | ||
| agent_response_callback=agent_response_callback, | ||
| ) | ||
|
|
||
| # 2. Create a runtime and start it | ||
| runtime = InProcessRuntime() | ||
| runtime.start() | ||
|
|
||
| # 3. Invoke the orchestration with a task and the runtime | ||
| orchestration_result = await magentic_orchestration.invoke( | ||
| task=( | ||
| "I am preparing a report on the energy efficiency of different machine learning model architectures. " | ||
| "Compare the estimated training and inference energy consumption of ResNet-50, BERT-base, and GPT-2 " | ||
| "on standard datasets (e.g., ImageNet for ResNet, GLUE for BERT, WebText for GPT-2). " | ||
| "Then, estimate the CO2 emissions associated with each, assuming training on an Azure Standard_NC6s_v3 VM " | ||
| "for 24 hours. Provide tables for clarity, and recommend the most energy-efficient model " | ||
| "per task type (image classification, text classification, and text generation)." | ||
| ), | ||
| runtime=runtime, | ||
| ) | ||
|
|
||
| # 4. Wait for the results | ||
| value = await orchestration_result.get() | ||
|
|
||
| print(f"\nFinal result:\n{value}") | ||
|
|
||
| # 5. Stop the runtime when idle | ||
| await runtime.stop_when_idle() | ||
|
|
||
| """ | ||
| Sample output: | ||
| **ResearchAgent** | ||
| Estimating the energy consumption and associated CO₂ emissions for training and inference of ResNet-50, BERT-base... | ||
TaoChenOSU marked this conversation as resolved.
Show resolved
Hide resolved
|
||
|
|
||
| **CoderAgent** | ||
| Here is the comparison of energy consumption and CO₂ emissions for each model (ResNet-50, BERT-base, and GPT-2) | ||
| over a 24-hour period: | ||
|
|
||
| | Model | Training Energy (kWh) | Inference Energy (kWh) | Total Energy (kWh) | CO₂ Emissions (kg) | | ||
| |-----------|------------------------|------------------------|---------------------|---------------------| | ||
| | ResNet-50 | 21.11 | 0.08232 | 21.19232 | 19.50 | | ||
| | BERT-base | 0.048 | 0.23736 | 0.28536 | 0.26 | | ||
| | GPT-2 | 42.22 | 0.35604 | 42.57604 | 39.17 | | ||
|
|
||
| ### Recommendations: | ||
| ... | ||
|
|
||
| **CoderAgent** | ||
| Here are the recalibrated results for energy consumption and CO₂ emissions, assuming a more conservative approach | ||
| for models like GPT-2: | ||
|
|
||
| | Model | Training Energy (kWh) | Inference Energy (kWh) | Total Energy (kWh) | CO₂ Emissions (kg) | | ||
| |------------------|------------------------|------------------------|---------------------|---------------------| | ||
| | ResNet-50 | 21.11 | 0.08232 | 21.19232 | 19.50 | | ||
| | BERT-base | 0.048 | 0.23736 | 0.28536 | 0.26 | | ||
| | GPT-2 (Adjusted) | 42.22 | 0.35604 | 42.57604 | 39.17 | | ||
|
|
||
| ... | ||
|
|
||
| **ResearchAgent** | ||
| Estimating the energy consumption and associated CO₂ emissions for training and inference of machine learning ... | ||
|
|
||
| **ResearchAgent** | ||
| Estimating the energy consumption and CO₂ emissions of training and inference for ResNet-50, BERT-base, and ... | ||
|
|
||
| **CoderAgent** | ||
| Here is the estimated energy use and CO₂ emissions for a full day of operation for each model on an Azure ... | ||
|
|
||
| **ResearchAgent** | ||
| Recent analyses have highlighted the substantial energy consumption and carbon emissions associated with ... | ||
|
|
||
| **CoderAgent** | ||
| Here's the refined estimation for the energy use and CO₂ emissions for optimized models on an Azure ... | ||
|
|
||
| **CoderAgent** | ||
| To provide precise estimates for CO₂ emissions based on Azure's regional data centers' carbon intensity, we need ... | ||
|
|
||
| **ResearchAgent** | ||
| To refine the CO₂ emission estimates for training and inference of ResNet-50, BERT-base, and GPT-2 on an Azure ... | ||
|
|
||
| **CoderAgent** | ||
| Here's the refined comparative table for energy consumption and CO₂ emissions for ResNet-50, BERT-base, and GPT-2, | ||
| taking into account carbon intensity data for Azure's West Europe and Sweden Central regions: | ||
|
|
||
| | Model | Energy (kWh) | CO₂ Emissions West Europe (kg) | CO₂ Emissions Sweden Central (kg) | | ||
| |------------|--------------|--------------------------------|-----------------------------------| | ||
| | ResNet-50 | 5.76 | 0.639 | 0.086 | | ||
| | BERT-base | 9.18 | 1.019 | 0.138 | | ||
| | GPT-2 | 12.96 | 1.439 | 0.194 | | ||
|
|
||
| **Refined Recommendations:** | ||
|
|
||
| ... | ||
|
|
||
| Final result: | ||
| Here is the comprehensive report on energy efficiency and CO₂ emissions for ResNet-50, BERT-base, and GPT-2 models | ||
| when trained and inferred on an Azure Standard_NC6s_v3 VM for 24 hours. | ||
|
|
||
| ### Energy Consumption and CO₂ Emissions: | ||
|
|
||
| Based on refined analyses, here are the estimated energy consumption and CO₂ emissions for each model: | ||
|
|
||
| | Model | Energy (kWh) | CO₂ Emissions West Europe (kg) | CO₂ Emissions Sweden Central (kg) | | ||
| |------------|--------------|--------------------------------|-----------------------------------| | ||
| | ResNet-50 | 5.76 | 0.639 | 0.086 | | ||
| | BERT-base | 9.18 | 1.019 | 0.138 | | ||
| | GPT-2 | 12.96 | 1.439 | 0.194 | | ||
|
|
||
| ### Recommendations for Energy Efficiency: | ||
|
|
||
| ... | ||
| """ | ||
|
|
||
|
|
||
| if __name__ == "__main__": | ||
| asyncio.run(main()) | ||
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Oops, something went wrong.
Oops, something went wrong.
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Uh oh!
There was an error while loading. Please reload this page.