Skip to main content

tatolab-stream

The Python SDK for Tatolab, which turns cameras, microphones, screens, models and your own code into realtime streams that browsers, other streams and AI agents can read.

Looking to get started? Install the Tatolab app first, not this package. tatolab-stream is the library you write streams with. It doesn't run them: the Tatolab runtime on your machine does, and it comes with the app.

  • macOS (Apple silicon, macOS 26 or later): brew install --cask tatolab/tap/tatolab, then open -a Tatolab
  • Linux (x86_64 with a Vulkan GPU): curl -fsSL https://get.tatolab.com/install.sh | sh

Then follow the Quickstart. tatolab new creates a project that already depends on this package, so you rarely pip install it yourself.

What it is

A stream is an ordinary Python project. You write nodes as Python classes and connect them in a function decorated @graph_generator:

from tatolab.camera import CameraSource
from tatolab.stream import StreamBuilder, graph_generator

from nodes.halftone_lens_effect import HalftoneLensEffect


@graph_generator
def main(stream_builder: StreamBuilder) -> None:
    camera = stream_builder.add(CameraSource)
    effect = stream_builder.add(HalftoneLensEffect)
    stream_builder.connect(camera.output("video"), effect.input("video_from_upstream"))

CameraSource comes from the tatolab-camera module, a dependency the project adds like any other. tatolab dev runs it in the runtime. Each node runs in its own process, so one slow node never blocks another. Tato Studio's viewer shows any node's output on the machine, and the tatolab-moq gateway nodes send one to a MoQ relay, where other streams and machines read it.

A generator may add another generator as it adds a node class, its keyword arguments the settings the inner generator promotes. The inner graph is stored as a fragment, whose GraphInput and GraphOutput nodes are its ports, and it runs flattened under its path of names (look/soft/blur).

This package holds the authoring API (@graph_generator, @node.process, StreamBuilder), typed classes for every built-in node (the converters between the Image, Array, Channels and Records families), and the protocols your nodes are handed at run time. It installs and imports without the runtime, with complete type hints, so your editor, type checker and unit tests work on any machine.

Add it to an existing project

uv add tatolab-stream

The runtime runs your nodes with the project's .venv/bin/python, so anything you install there your nodes can import. See Python SDK for details.

Supports CPython 3.12 to 3.13 (not the free-threaded builds).

Other packages

  • tatolab-webrtc: WHIP publish and WHEP play nodes.
  • tatolab on PyPI is a placeholder that installs nothing. Tatolab itself installs as an app, as above.

Licensed under the Business Source License 1.1.

Metadata

Release files for tatolab-stream 0.68.98

For a detailed explanation of source distributions (sdists) and built distributions (wheels), please see the package formats documentation.

Source distribution (sdist)

Source distribution for tatolab-stream 0.68.98
File Size Uploaded Provenance Actions
tatolab_stream-0.68.98.tar.gz 122.1 kB Without provenance Details for tatolab_stream-0.68.98.tar.gz

Built distribution (wheel)

Table of built distributions (wheels) for tatolab-stream 0.68.98
File Interpreter ABI Platform Actions
tatolab_stream-0.68.98-py3-none-any.whl Python 3 none any Details for tatolab_stream-0.68.98-py3-none-any.whl

Total release size: 263.6 kB

Release files / tatolab_stream-0.68.98.tar.gz

Download URL tatolab_stream-0.68.98.tar.gz
Size 122.1 kB
Tags Source
SHA-256 checksum
How to use checksums
99c907e958139aa7b879e0e475332de76ded97cc1542c54a75802371115b2847
BLAKE2b-256 checksum
How to use checksums
e2389b2b34bef495511a7568357eab415f21f1732133f4b1affa328d2afcd930
Upload date
Uploaded using Trusted Publishing?
What is trusted publishing?
No
Uploaded via uv/0.12.23 {"installer":{"name":"uv","version":"0.12.23","subcommand":["publish"]},"python":null,"implementation":{"name":null,"version":null},"distro":{"name":"Ubuntu","version":"26.04","id":"resolute","libc":null},"system":{"name":null,"release":null},"cpu":null,"openssl_version":null,"setuptools_version":null,"rustc_version":null,"ci":true}

Release files / tatolab_stream-0.68.98-py3-none-any.whl

Download URL tatolab_stream-0.68.98-py3-none-any.whl
Size 141.5 kB
Tags Python 3
SHA-256 checksum
How to use checksums
d054792fa6b382ffd9e6bd9bb4f01c38d75412ec721f865cf10a4df5843b10b0
BLAKE2b-256 checksum
How to use checksums
1eda5cc33d549c23c76fca73c0bf53b9a74e552bd98c2c40e24f1d21b0123397
Upload date
Uploaded using Trusted Publishing?
What is trusted publishing?
No
Uploaded via uv/0.12.23 {"installer":{"name":"uv","version":"0.12.23","subcommand":["publish"]},"python":null,"implementation":{"name":null,"version":null},"distro":{"name":"Ubuntu","version":"26.04","id":"resolute","libc":null},"system":{"name":null,"release":null},"cpu":null,"openssl_version":null,"setuptools_version":null,"rustc_version":null,"ci":true}

Security for tatolab-stream 0.68.98

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.

No provenance found

None of the files in this release were uploaded with provenance metadata.

A note on package security

PyPI is a repository for distributing packages, not a service for auditing them. While third-party security researchers and our staff do their best to detect and remove malicious packages, PyPI cannot guarantee the quality, security or suitability of any given package.

To learn more about protecting yourself from malicious actors, visit our package security documentation.

Release history Release notifications | RSS feed

This release

0.68.98 This release

2 release files No provenance

Anthropic, PBC Visionary sponsor Bloomberg Visionary sponsor Hudson River Trading Visionary sponsor Meta Visionary sponsor NVIDIA Visionary sponsor Microsoft Sustainability sponsor Depot Continuous Integration AWS Cloud computing and Security Sponsor Datadog Monitoring Fastly CDN Google Download Analytics Sentry Error logging StatusPage Status page