very-good-ffmpeg 1.0.2

Python SDK for the Very Good FFmpeg API

very-good-ffmpeg

Python SDK for the Very Good FFmpeg API — run FFmpeg jobs in the cloud.

Homepage · API Docs · GitHub

Installation

pip install very-good-ffmpeg

Usage

import os
from very_good_ffmpeg import VGF

client = VGF(os.environ["VGFFMPEG_API_KEY"])

# Create a job and wait for it to complete
job = client.run(
    input_files={"input": "https://example.com/input.mp4"},
    output_files=["output.mp4"],
    ffmpeg_commands=["-i {{input}} -vf scale=1280:720 {{output.mp4}}"],
    wait=True,
)

print(job.status)       # "succeeded"
print(job.output_files) # {"output.mp4": "https://..."}

Async is also supported — use arun, jobs.aget, jobs.alist, jobs.acancel, files.aupload, etc.

import asyncio, os
from very_good_ffmpeg import VGF

async def main():
    client = VGF(os.environ["VGFFMPEG_API_KEY"])
    job = await client.arun(
        input_files={"input": "https://example.com/input.mp4"},
        output_files=["output.mp4"],
        ffmpeg_commands=["-i {{input}} -vf scale=1280:720 {{output.mp4}}"],
        wait=True,
    )
    print(job.status)

asyncio.run(main())

Reference

Create a job — client.run(...) / client.arun(...)

Submits an FFmpeg job. Pass wait=True to block until the job reaches a terminal state.

job = client.run(
    input_files={"input": "https://example.com/input.mp4"},
    output_files=["output.mp4"],
    ffmpeg_commands=["-i {{input}} -c:v libx264 {{output.mp4}}"],
    webhook_url="https://example.com/webhook",  # optional
    machine="nvidia",  # optional: "cpu" | "nvidia"
    wait=True,
)

Get a job — client.jobs.get(id) / client.jobs.aget(id)

job = client.jobs.get("job_abc123")
print(job.status)  # "queued" | "running" | "succeeded" | "failed" | "cancelled"

List jobs — client.jobs.list(...) / client.jobs.alist(...)

Returns a JobList with data (list of jobs) and paging_params.

result = client.jobs.list(limit=20, offset=0)
print(result.data)          # list[Job]
print(result.paging_params) # PagingParams(limit=20, offset=0, total=100, has_more=True)

Cancel a job — client.jobs.cancel(id) / client.jobs.acancel(id)

job = client.jobs.cancel("job_abc123")
print(job.status)  # "cancelled"

Poll until done — job.wait(...) / job.async_wait(...)

job = client.run(input_files=..., output_files=..., ffmpeg_commands=...)
job.wait(timeout=120)
print(job.status)

Upload a file — client.files.upload(data, content_type?) / client.files.aupload(...)

Upload bytes or a file path to temporary storage and get back a URL you can use as an input_files value.

with open("input.mp4", "rb") as f:
    download_url = client.files.upload(f.read(), "video/mp4")

job = client.run(
    input_files={"input": download_url},
    output_files=["output.mp4"],
    ffmpeg_commands=["-i {{input}} -vf scale=1280:720 {{output.mp4}}"],
)

Pass a file path directly to have the SDK read it for you:

download_url = client.files.upload("input.mp4")

Prepare upload — client.files.prepare() / client.files.aprepare()

Returns a TmpFile with upload_url and download_url if you need to stream a large file yourself.

tmp = client.files.prepare()
# PUT your file to tmp.upload_url, then use tmp.download_url as input

Examples

See examples/example/main.py for a runnable example covering both sync and async usage.

cd examples/example
pip install -r requirements.txt
VGFFMPEG_API_KEY=your_key python main.py

Support

Open an issue at github.com/verygoodffmpeg/Very-Good-FFmpeg-Python-SDK or email [e-mail].