GitHub Repository

Embed, cluster, and visualize any collection of texts in 3D semantic space — then learn a continuous semantic flow field over that space, so you can see not just where texts are, but how meaning tends to move between them.

18 starsPython

A semantic flow tool for embeddings

Name: A semantic flow tool for embeddings
Availability: InStock
Author: pixedar

by pixedar·Apr 15, 2026·1 point·0 comments

Visit Project View on HN

AI Analysis

●●SolidBig BrainRabbit HoleWizardry

Flow fields over embedding trajectories—TensorFlow Projector but with dynamics.

Strengths

•Continuous flow field reveals how meaning drifts between clusters over time.
•RBF and MDN flow models let users trade speed against local structure fidelity.
•Works on diverse data: chat logs, agent traces, research papers, support tickets.

Weaknesses

•Experimental alpha with 1 star—needs real user testing and documentation.
•PyTorch dependency for MDN model adds friction for lightweight deployments.

Post Description

So there are a ton of tools that let you visualize trajectories in semantic space but I just had the idea to compute the flow field out of these trajectories

The story behind it was that I initially analyzed human emotions over time in terms of repeating behavior patterns, and from that I had the idea to generalize it to any data. It’s very experimental alpha at this stage