Fal

FalTools enable an Agent to perform media generation tasks.

Prerequisites

The following example requires the fal_client library and an API key which can be obtained from Fal.

1uv pip install -U fal_client

1export FAL_KEY=***

Example

The following agent will use FAL to generate any video requested by the user.

1from kern.agent import Agent
2from kern.models.openai import OpenAIResponses
3from kern.tools.fal import FalTools
4
5fal_agent = Agent(
6    name="Fal Video Generator Agent",
7    model=OpenAIResponses(id="gpt-5.2"),
8    tools=[FalTools("fal-ai/hunyuan-video")],
9    description="You are an AI agent that can generate videos using the Fal API.",
10    instructions=[
11        "When the user asks you to create a video, use the `generate_media` tool to create the video.",
12        "Return the URL as raw to the user.",
13        "Don't convert video URL to markdown or anything else.",
14    ],
15    markdown=True,
16    debug_mode=True,
17    )
18
19fal_agent.print_response("Generate video of balloon in the ocean")

Toolkit Params

Parameter	Type	Default	Description
`api_key`	`str`	`None`	API key for authentication purposes.
`model`	`str`	`None`	The model to use for the media generation.
`enable_generate_media`	`bool`	`True`	Enable the generate_media functionality.
`enable_image_to_image`	`bool`	`True`	Enable the image_to_image functionality.
`all`	`bool`	`False`	Enable all functionality.

Toolkit Functions

Function	Description
`generate_media`	Generate either images or videos depending on the user prompt.
`image_to_image`	Transform an input image based on a text prompt.

Developer Resources

View Tools