I've been building TokenGate - an experimental Python concurrency engine that uses a token-based model to manage threaded tasks. No manual thread management, no futures, no ThreadPoolExecutor. Just a 3 line coordinator and single line decorators to manage all threading.
It runs on Python 3.12 with the GIL effectively "gone".
On my machine (Ryzen 7800 / RTX 4070 Super) I'm seeing 7.25x concurrency across 8 tasks of mixed work and 6.01x on sustained high variety workloads - but that's just one setup. I want to know what it does on yours.
What I'm asking
Try the demos (or make an app!), paste your results, that's it.
The whole demo suite takes about 5 minutes to set up and the results
tell me a lot about how the engine scales across different hardware.
How to get started
Option 1 - Direct download (fastest):
Grab the beta zip from tavari.online
Option 2 - Clone the repo:
git clone https://github.com/TavariAgent/Py-TokenGate
cd Py-TokenGate
pip install -r requirements.txt
Then check BETA.md
for the quick start.
What TokenGate actually does
- Decorates synchronous functions with
@task_token_guard- one line, done - Routes tasks through a token-managed thread pool automatically
- Built-in DoS protection to prevent the system from overwhelming itself
- Live telemetry via WebSocket GUI at
localhost:5000 - Works standalone or with the WebSocket dashboard
What I want to hear back
- Your hardware (CPU model, core count)
- Your experience (how did it work for you?)
Drop results here, open an Issue on GitHub or leave me feedback on tavari.online.
This is an active beta - rough edges are expected.
I'm self-taught, this is the work my learning experience produced.
If it's useful or interesting to you, I appreciate the feedback.