MiniMax Tools
Demonstrates MiniMax video generation tools.
"""
MiniMax Tools
=============================
Demonstrates MiniMax video generation tools.
The toolkit submits a text-to-video task to the MiniMax video generation API,
polls it until it reaches a terminal state, and returns the result as a Video
artifact.
Setup:
export MINIMAX_API_KEY=***
Prompts to try:
- "Generate a video of a paper boat crossing a moonlit lake"
- "Create a 10 second video of a hot air balloon over a desert at sunrise"
"""
from agno.agent import Agent
from agno.models.openai import OpenAIChat
from agno.tools.minimax import MiniMaxTools
# ---------------------------------------------------------------------------
# Create Agent
# ---------------------------------------------------------------------------
minimax_agent = Agent(
name="MiniMax Video Generator Agent",
model=OpenAIChat(id="gpt-5.6-luna"),
tools=[
MiniMaxTools(
# Use region="cn_zh" for the mainland China endpoint.
region="global_en",
model="MiniMax-H3",
)
],
description="You are an AI agent that can generate videos using the MiniMax API.",
instructions=[
"When the user asks you to create a video, use the `generate_video` tool.",
"Duration is given in whole seconds, from 4 through 15.",
"Return the URL as raw to the user.",
"Don't convert video URL to markdown or anything else.",
],
markdown=True,
)
# ---------------------------------------------------------------------------
# Run Agent
# ---------------------------------------------------------------------------
if __name__ == "__main__":
minimax_agent.print_response(
"Generate a video of a paper boat crossing a moonlit lake"
)
# ---------------------------------------------------------------------------
# More Examples
# ---------------------------------------------------------------------------
# Vertical 9:16 output at a specific duration.
# minimax_agent.print_response(
# "Create a 10 second vertical video of a hot air balloon over a desert at sunrise"
# )The toolkit submits a remote job and polls for up to 600 seconds by default. It returns a Video artifact after a successful result; a timeout is not a remote cancellation. To obtain the URL without relying on the model's final wording, replace the final call with:
response = minimax_agent.run("Generate a video of a paper boat crossing a moonlit lake")
for video in response.videos or []:
print(video.url)region selects the endpoint and must match your provider account. Duration and resolution are sent to MiniMax for validation.
Run the Example
Set up your virtual environment
uv venv --python 3.12
source .venv/bin/activateInstall dependencies
uv pip install -U agno openaiExport your API keys
export MINIMAX_API_KEY="your_minimax_api_key_here"
export OPENAI_API_KEY="your_openai_api_key_here"Run the example
Save the code above as minimax_tools.py, then run:
python minimax_tools.pyFull source: cookbook/91_tools/minimax_tools.py