-
Notifications
You must be signed in to change notification settings - Fork 3k
Closed
Labels
questionQuestion about using the SDKQuestion about using the SDK
Description
Please read this first
- Have you read the docs?Agents SDK docs - Yes
- Have you searched for related issues? Others may have had similar requests - Yes
Question
def foo(a):
'''Test function foo'''
return {ToolOutputImage(image_url="https://tinyurl.com/3ut4a4fn")}
the docs talks about Returning images or files from function tools using ToolOutputImage. When I do this, the LLM(gemini) is unable to read the image.
This is my workflow
client = AsyncOpenAI(
base_url="https://generativelanguage.googleapis.com/v1beta/openai/",
api_key=GEMINI_API_KEY
)
set_default_openai_client(client)
set_tracing_disabled(True)
# Define the Model
model = OpenAIChatCompletionsModel(
model="gemini-2.5-flash",
openai_client=client
)
# Create the Agent with Tools
stock_analysis_agent = Agent(
name="TestAgent",
instructions="You are a helpful agent",
tools=[function_tool(foo), function_tool(foo2), function_tool(foo3)],
model=model
)
async def main(prompt):
result = await Runner.run(stock_analysis_agent, prompt)
return result.final_output
Metadata
Metadata
Assignees
Labels
questionQuestion about using the SDKQuestion about using the SDK