Skip to content

A programming framework for agentic AI

Add gpt-4o-mini, update examples (#253)

Eric Zhu176932ef9cab
diff --git a/python/docs/src/core-concepts/ai-agents.md b/python/docs/src/core-concepts/ai-agents.md
index 006ea850c..2177fe853 100644
--- a/python/docs/src/core-concepts/ai-agents.md
+++ b/python/docs/src/core-concepts/ai-agents.md
@@ -71,7 +71,7 @@ async def main():
runtime = SingleThreadedAgentRuntime()
agent = await runtime.register_and_get(
"simple-agent",
- lambda: SimpleAgent(OpenAIChatCompletionClient(model="gpt-3.5-turbo", api_key="YOUR_API_KEY")),
+ lambda: SimpleAgent(OpenAIChatCompletionClient(model="gpt-4o-mini", api_key="YOUR_API_KEY")),
# Leave out the api_key field if you have the API key in the environment variable OPENAI_API_KEY
)
# Start the runtime processing messages.
@@ -257,7 +257,7 @@ async def main() -> None:
agent = await runtime.register_and_get(
"tool-agent",
lambda: ToolAgent(
- OpenAIChatCompletionClient(model="gpt-3.5-turbo", api_key="YOUR_API_KEY"),
+ OpenAIChatCompletionClient(model="gpt-4o-mini", api_key="YOUR_API_KEY"),
tools=[
FunctionTool(get_stock_price, description="Get the stock price."),
],
diff --git a/python/samples/core/one_agent_direct.py b/python/samples/core/one_agent_direct.py
index ec8b298ee..b7b7b9a56 100644
--- a/python/samples/core/one_agent_direct.py
+++ b/python/samples/core/one_agent_direct.py
@@ -47,7 +47,7 @@ async def main() -> None:
runtime = SingleThreadedAgentRuntime()
agent = await runtime.register_and_get(
"chat_agent",
- lambda: ChatCompletionAgent("Chat agent", get_chat_completion_client_from_envs(model="gpt-3.5-turbo")),
+ lambda: ChatCompletionAgent("Chat agent", get_chat_completion_client_from_envs(model="gpt-4o-mini")),
)
run_context = runtime.start()
diff --git a/python/samples/core/two_agents_pub_sub.py b/python/samples/core/two_agents_pub_sub.py
index ffaba2c17..e33db7d3d 100644
--- a/python/samples/core/two_agents_pub_sub.py
+++ b/python/samples/core/two_agents_pub_sub.py
@@ -81,7 +81,7 @@ async def main() -> None:
"Jack",
lambda: ChatCompletionAgent(
description="Jack a comedian",
- model_client=get_chat_completion_client_from_envs(model="gpt-3.5-turbo"),
+ model_client=get_chat_completion_client_from_envs(model="gpt-4o-mini"),
system_messages=[
SystemMessage("You are a comedian likes to make jokes. " "When you are done talking, say 'TERMINATE'.")
],
@@ -92,7 +92,7 @@ async def main() -> None:
"Cathy",
lambda: ChatCompletionAgent(
description="Cathy a poet",
- model_client=get_chat_completion_client_from_envs(model="gpt-3.5-turbo"),
+ model_client=get_chat_completion_client_from_envs(model="gpt-4o-mini"),
system_messages=[
SystemMessage("You are a poet likes to write poems. " "When you are done talking, say 'TERMINATE'.")
],
diff --git a/python/samples/patterns/coder_reviewer.py b/python/samples/patterns/coder_reviewer.py
index acddc5f38..77be82f64 100644
--- a/python/samples/patterns/coder_reviewer.py
+++ b/python/samples/patterns/coder_reviewer.py
@@ -255,14 +255,14 @@ async def main() -> None:
"ReviewerAgent",
lambda: ReviewerAgent(
description="Code Reviewer",
- model_client=get_chat_completion_client_from_envs(model="gpt-3.5-turbo"),
+ model_client=get_chat_completion_client_from_envs(model="gpt-4o-mini"),
),
)
await runtime.register(
"CoderAgent",
lambda: CoderAgent(
description="Coder",
- model_client=get_chat_completion_client_from_envs(model="gpt-3.5-turbo"),
+ model_client=get_chat_completion_client_from_envs(model="gpt-4o-mini"),
),
)
run_context = runtime.start()
diff --git a/python/samples/patterns/group_chat.py b/python/samples/patterns/group_chat.py
index 710a7c6ed..7f4962d11 100644
--- a/python/samples/patterns/group_chat.py
+++ b/python/samples/patterns/group_chat.py
@@ -118,7 +118,7 @@ async def main() -> None:
lambda: GroupChatParticipant(
description="A data scientist",
system_messages=[SystemMessage("You are a data scientist.")],
- model_client=get_chat_completion_client_from_envs(model="gpt-3.5-turbo"),
+ model_client=get_chat_completion_client_from_envs(model="gpt-4o-mini"),
),
)
agent2 = await runtime.register_and_get(
@@ -126,7 +126,7 @@ async def main() -> None:
lambda: GroupChatParticipant(
description="An engineer",
system_messages=[SystemMessage("You are an engineer.")],
- model_client=get_chat_completion_client_from_envs(model="gpt-3.5-turbo"),
+ model_client=get_chat_completion_client_from_envs(model="gpt-4o-mini"),
),
)
agent3 = await runtime.register_and_get(
@@ -134,7 +134,7 @@ async def main() -> None:
lambda: GroupChatParticipant(
description="An artist",
system_messages=[SystemMessage("You are an artist.")],
- model_client=get_chat_completion_client_from_envs(model="gpt-3.5-turbo"),
+ model_client=get_chat_completion_client_from_envs(model="gpt-4o-mini"),
),
)
diff --git a/python/samples/patterns/mixture_of_agents.py b/python/samples/patterns/mixture_of_agents.py
index 4ef5e3e4a..4b1f5c073 100644
--- a/python/samples/patterns/mixture_of_agents.py
+++ b/python/samples/patterns/mixture_of_agents.py
@@ -117,7 +117,7 @@ async def main() -> None:
lambda: ReferenceAgent(
description="Reference Agent 1",
system_messages=[SystemMessage("You are a helpful assistant that can answer questions.")],
- model_client=get_chat_completion_client_from_envs(model="gpt-3.5-turbo", temperature=0.1),
+ model_client=get_chat_completion_client_from_envs(model="gpt-4o-mini", temperature=0.1),
),
)
await runtime.register(
@@ -125,7 +125,7 @@ async def main() -> None:
lambda: ReferenceAgent(
description="Reference Agent 2",
system_messages=[SystemMessage("You are a helpful assistant that can answer questions.")],
- model_client=get_chat_completion_client_from_envs(model="gpt-3.5-turbo", temperature=0.5),
+ model_client=get_chat_completion_client_from_envs(model="gpt-4o-mini", temperature=0.5),
),
)
await runtime.register(
@@ -133,7 +133,7 @@ async def main() -> None:
lambda: ReferenceAgent(
description="Reference Agent 3",
system_messages=[SystemMessage("You are a helpful assistant that can answer questions.")],
- model_client=get_chat_completion_client_from_envs(model="gpt-3.5-turbo", temperature=1.0),
+ model_client=get_chat_completion_client_from_envs(model="gpt-4o-mini", temperature=1.0),
),
)
await runtime.register(
@@ -145,7 +145,7 @@ async def main() -> None:
"...synthesize these responses into a single, high-quality response... Responses from models:"
)
],
- model_client=get_chat_completion_client_from_envs(model="gpt-3.5-turbo"),
+ model_client=get_chat_completion_client_from_envs(model="gpt-4o-mini"),
num_references=3,
),
)
diff --git a/python/samples/patterns/multi_agent_debate.py b/python/samples/patterns/multi_agent_debate.py
index 57d3baed7..e99805bb7 100644
--- a/python/samples/patterns/multi_agent_debate.py
+++ b/python/samples/patterns/multi_agent_debate.py
@@ -214,7 +214,7 @@ async def main(question: str) -> None:
await runtime.register(
"MathSolver1",
lambda: MathSolver(
- get_chat_completion_client_from_envs(model="gpt-3.5-turbo"),
+ get_chat_completion_client_from_envs(model="gpt-4o-mini"),
neighbor_names=["MathSolver2", "MathSolver4"],
max_round=3,
),
@@ -222,7 +222,7 @@ async def main(question: str) -> None:
await runtime.register(
"MathSolver2",
lambda: MathSolver(
- get_chat_completion_client_from_envs(model="gpt-3.5-turbo"),
+ get_chat_completion_client_from_envs(model="gpt-4o-mini"),
neighbor_names=["MathSolver1", "MathSolver3"],
max_round=3,
),
@@ -230,7 +230,7 @@ async def main(question: str) -> None:
await runtime.register(
"MathSolver3",
lambda: MathSolver(
- get_chat_completion_client_from_envs(model="gpt-3.5-turbo"),
+ get_chat_completion_client_from_envs(model="gpt-4o-mini"),
neighbor_names=["MathSolver2", "MathSolver4"],
max_round=3,
),
@@ -238,7 +238,7 @@ async def main(question: str) -> None:
await runtime.register(
"MathSolver4",
lambda: MathSolver(
- get_chat_completion_client_from_envs(model="gpt-3.5-turbo"),
+ get_chat_completion_client_from_envs(model="gpt-4o-mini"),
neighbor_names=["MathSolver1", "MathSolver3"],
max_round=3,
),
diff --git a/python/samples/tool-use/coding_one_agent_direct.py b/python/samples/tool-use/coding_one_agent_direct.py
index 9facda9fc..3b029d354 100644
--- a/python/samples/tool-use/coding_one_agent_direct.py
+++ b/python/samples/tool-use/coding_one_agent_direct.py
@@ -134,7 +134,7 @@ async def main() -> None:
lambda: ToolEnabledAgent(
description="Tool Use Agent",
system_messages=[SystemMessage("You are a helpful AI Assistant. Use your tools to solve problems.")],
- model_client=get_chat_completion_client_from_envs(model="gpt-3.5-turbo"),
+ model_client=get_chat_completion_client_from_envs(model="gpt-4o-mini"),
tools=tools,
),
)
diff --git a/python/samples/tool-use/coding_one_agent_direct_intercept.py b/python/samples/tool-use/coding_one_agent_direct_intercept.py
index a7bdc7a67..be26b96c1 100644
--- a/python/samples/tool-use/coding_one_agent_direct_intercept.py
+++ b/python/samples/tool-use/coding_one_agent_direct_intercept.py
@@ -52,7 +52,7 @@ async def main() -> None:
lambda: ToolEnabledAgent(
description="Tool Use Agent",
system_messages=[SystemMessage("You are a helpful AI Assistant. Use your tools to solve problems.")],
- model_client=get_chat_completion_client_from_envs(model="gpt-3.5-turbo"),
+ model_client=get_chat_completion_client_from_envs(model="gpt-4o-mini"),
tools=tools,
),
)
diff --git a/python/samples/tool-use/coding_two_agent_pub_sub.py b/python/samples/tool-use/coding_two_agent_pub_sub.py
index c9ba45cb0..61d4719aa 100644
--- a/python/samples/tool-use/coding_two_agent_pub_sub.py
+++ b/python/samples/tool-use/coding_two_agent_pub_sub.py
@@ -197,7 +197,7 @@ async def main() -> None:
lambda: ToolUseAgent(
description="Tool Use Agent",
system_messages=[SystemMessage("You are a helpful AI Assistant. Use your tools to solve problems.")],
- model_client=get_chat_completion_client_from_envs(model="gpt-3.5-turbo"),
+ model_client=get_chat_completion_client_from_envs(model="gpt-4o-mini"),
tools=tools,
),
)
diff --git a/python/samples/tool-use/custom_function_tool_one_agent_direct.py b/python/samples/tool-use/custom_function_tool_one_agent_direct.py
index ce0af1206..156bffa21 100644
--- a/python/samples/tool-use/custom_function_tool_one_agent_direct.py
+++ b/python/samples/tool-use/custom_function_tool_one_agent_direct.py
@@ -37,7 +37,7 @@ async def main() -> None:
lambda: ToolEnabledAgent(
description="Tool Use Agent",
system_messages=[SystemMessage("You are a helpful AI Assistant. Use your tools to solve problems.")],
- model_client=get_chat_completion_client_from_envs(model="gpt-3.5-turbo"),
+ model_client=get_chat_completion_client_from_envs(model="gpt-4o-mini"),
tools=[
# Define a tool that gets the stock price.
FunctionTool(
diff --git a/python/src/agnext/components/models/_model_info.py b/python/src/agnext/components/models/_model_info.py
index 06a28d6ad..583765a52 100644
--- a/python/src/agnext/components/models/_model_info.py
+++ b/python/src/agnext/components/models/_model_info.py
@@ -6,6 +6,7 @@ from ._model_client import ModelCapabilities
# This is a moving target, so correctness is checked by the model value returned by openai against expected values at runtime``
_MODEL_POINTERS = {
"gpt-4o": "gpt-4o-2024-05-13",
+ "gpt-4o-mini": "gpt-4o-mini-2024-07-18",
"gpt-4-turbo": "gpt-4-turbo-2024-04-09",
"gpt-4-turbo-preview": "gpt-4-0125-preview",
"gpt-4": "gpt-4-0613",
@@ -20,6 +21,11 @@ _MODEL_CAPABILITIES: Dict[str, ModelCapabilities] = {
"function_calling": True,
"json_output": True,
},
+ "gpt-4o-mini-2024-07-18": {
+ "vision": True,
+ "function_calling": True,
+ "json_output": True,
+ },
"gpt-4-turbo-2024-04-09": {
"vision": True,
"function_calling": True,
@@ -79,6 +85,7 @@ _MODEL_CAPABILITIES: Dict[str, ModelCapabilities] = {
_MODEL_TOKEN_LIMITS: Dict[str, int] = {
"gpt-4o-2024-05-13": 128000,
+ "gpt-4o-mini-2024-07-18": 128000,
"gpt-4-turbo-2024-04-09": 128000,
"gpt-4-0125-preview": 128000,
"gpt-4-1106-preview": 128000,