Decart Python SDK
A Python SDK for Decart's models.
Installation
Using UV
uv add decart
Using pip
pip install decart
Documentation
For complete documentation, guides, and examples, visit: https://docs.platform.decart.ai/sdks/python
Quick Start
Image Editing (Process API)
import asyncio
import os
from decart import DecartClient, models
async def main():
async with DecartClient(api_key=os.getenv("DECART_API_KEY")) as client:
# Edit an image
result = await client.process({
"model": models.image("lucy-image-2"),
"prompt": "Apply a painterly oil-on-canvas look while preserving the composition",
"data": open("input.png", "rb"),
})
with open("output.png", "wb") as f:
f.write(result)
asyncio.run(main())
Video Editing (Queue API)
For video editing jobs, use the queue API to submit jobs and poll for results:
async with DecartClient(api_key=os.getenv("DECART_API_KEY")) as client:
# Submit and poll automatically
result = await client.queue.submit_and_poll({
"model": models.video("lucy-clip"),
"prompt": "Restyle this footage with anime shading and vibrant neon highlights",
"data": open("input.mp4", "rb"),
"on_status_change": lambda job: print(f"Status: {job.status}"),
})
if result.status == "completed":
with open("output.mp4", "wb") as f:
f.write(result.data)
else:
print(f"Job failed: {result.error}")
Or manage the polling manually:
async with DecartClient(api_key=os.getenv("DECART_API_KEY")) as client:
# Submit the job
job = await client.queue.submit({
"model": models.video("lucy-clip"),
"prompt": "Add cinematic teal-and-orange grading and gentle film grain",
"data": open("input.mp4", "rb"),
})
print(f"Job ID: {job.job_id}")
# Poll for status
status = await client.queue.status(job.job_id)
print(f"Status: {status.status}")
# Get result when completed
if status.status == "completed":
data = await client.queue.result(job.job_id)
with open("output.mp4", "wb") as f:
f.write(data)
Realtime fast mode
Realtime sessions accept an optional speed on RealtimeConnectOptions, alongside resolution.
Fast mode (speed="fast") serves the session from a higher-compute tier for lower latency and
higher throughput; output quality is unchanged. It is currently available for lucy-2.5 /
lucy-latest and lucy-vton-3.5 / lucy-vton-latest, in the US region only, and is billed at
2x the standard realtime rate for those models. Other models ignore the option (the SDK emits a
warning). Omit it (the default) for standard mode.
from decart import DecartClient, models
from decart.realtime import RealtimeClient, RealtimeConnectOptions
client = DecartClient(api_key=os.getenv("DECART_API_KEY"))
realtime = await RealtimeClient.connect(
base_url=client.realtime_base_url,
api_key=client.api_key,
local_track=local_track,
options=RealtimeConnectOptions(
model=models.realtime("lucy-2.5"),
on_remote_stream=on_remote_stream,
speed="fast", # omit for standard mode
),
)
Each model definition lists the speed tiers it advertises via ModelDefinition.supported_speeds
(for example models.realtime("lucy-2.5").supported_speeds == ("fast",)). See the
realtime docs for the full realtime API.
Development
Setup with UV
# Clone the repository
git clone https://github.com/decartai/decart-python
cd decart-python
# Install UV
curl -LsSf https://astral.sh/uv/install.sh | sh
# Install all dependencies (including dev dependencies)
uv sync --all-extras
# Run tests
uv run pytest
# Run linting
uv run ruff check decart/ tests/ examples/
# Format code
uv run black decart/ tests/ examples/
# Type check
uv run mypy decart/
Common Commands
# Install dependencies
uv sync --all-extras
# Run tests with coverage
uv run pytest --cov=decart --cov-report=html
# Run examples
uv run python examples/process_video.py
uv run python examples/realtime_synthetic.py
# Update dependencies
uv lock --upgrade
Test UI
The SDK includes an interactive test UI built with Gradio for quickly testing all SDK features without writing code.
# Install Gradio
pip install gradio
# Run the test UI
python test_ui.py
Then open http://localhost:7860 in your browser.
The UI provides tabs for:
- Image Editing - Image-to-image edits
- Video Editing - Video-to-video edits
- Video Restyle - Restyle videos using text prompts or reference images
- Tokens - Create short-lived client tokens
Enter your API key at the top of the interface to start testing.
Publishing a New Version
The package is automatically published to PyPI when you create a GitHub release.
Automated Release
Use the release script to automate the entire process:
python release.py
The script will:
- Display the current version
- Prompt for the new version
- Update
pyproject.toml - Commit and push changes
- Create a GitHub release with release notes
The GitHub Actions workflow will automatically build, test, and publish to PyPI.
License
MIT
1 maintainer
Metadata
1 maintainer
Release files for decart 0.2.1
For a detailed explanation of source distributions (sdists) and built distributions (wheels), please see the package formats documentation.
Source distribution (sdist)
| File | Size | Uploaded | |
|---|---|---|---|
| decart-0.2.1.tar.gz | 166.5 kB | Details |
Built distribution (wheel)
| File | Interpreter | ABI | Platform | Reset |
|---|---|---|---|---|
| decart-0.2.1-py3-none-any.whl | Python 3 | none | any | Details |
Total release size: 202.8 kB
Release files / decart-0.2.1.tar.gz
| Download URL | decart-0.2.1.tar.gz |
|---|---|
| Size | 166.5 kB |
| Tags | Source |
|
SHA-256 checksum How to use checksums |
9e36796b9d6993b57bd4d410e1db3246affce4b6e96e2eeb24ad1a0e8538c70b
|
|
BLAKE2b-256 checksum How to use checksums |
599a80ebc5ede25aae3a304f774213d38d4c9b27dbc2293c1a61da256c36318e
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
Yes |
| Uploaded via |
twine/7.0.0 CPython/3.13.14
|
Provenance
Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.
PyPI Publish Attestation
PyPI verified that this artifact, at this checksum, originated from the publisher listed below.
Signed by GitHub Actions, verified by PyPI on Sep 24, 2026.
Transparency logRelease files / decart-0.2.1-py3-none-any.whl
| Download URL | decart-0.2.1-py3-none-any.whl |
|---|---|
| Size | 36.4 kB |
| Tags | Python 3 |
|
SHA-256 checksum How to use checksums |
311029efa757811f581245c3e73667f582913661a46868958a86f678c55af690
|
|
BLAKE2b-256 checksum How to use checksums |
a265b1ddf3eae63ae001f710f52b8d382738035bfc9c4b36424e517a6e588ad6
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
Yes |
| Uploaded via |
twine/7.0.0 CPython/3.13.14
|
Provenance
Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.
PyPI Publish Attestation
PyPI verified that this artifact, at this checksum, originated from the publisher listed below.
Signed by GitHub Actions, verified by PyPI on Sep 24, 2026.
Transparency log
