# Machine learning visualization tools

> Training metrics, experiment tracking, embeddings and LLM traces. 15 tools, ranked by GitHub stars.

Tools for visualizing machine learning work: training metrics and experiment tracking, model debugging, embeddings, and traces of LLM applications. Most are Python packages that log from training or inference code to a local or hosted dashboard.

| # | Tool | GitHub stars | Commits (12 mo) | License | Description |
|---:|---|---:|---:|---|---|
| 1 | [Netron](https://awesomedataviz.com/tools/netron/) | 33,544 | 926 | MIT | Viewer for neural network, deep learning and machine learning models. |
| 2 | [PlotNeuralNet](https://awesomedataviz.com/tools/plotneuralnet/) | 25,009 | 0 | MIT | LaTeX code for drawing neural network architecture diagrams. |
| 3 | [Opik](https://awesomedataviz.com/tools/opik/) | 22,373 | 4,323 | Apache-2.0 | Formerly CometLLM. Debug, evaluate, and monitor LLM applications with tracing and dashboards. |
| 4 | [Phoenix](https://awesomedataviz.com/tools/phoenix/) | 11,704 | 3,725 | Elastic-2.0 | ML observability in a notebook with UMAP visualizations |
| 5 | [FiftyOne](https://awesomedataviz.com/tools/fiftyone/) | 11,143 | 8,917 | Apache-2.0 | Tool for visualizing, curating and evaluating computer vision datasets and models. |
| 6 | [Visdom](https://awesomedataviz.com/tools/visdom/) | 10,318 | 548 | Apache-2.0 | Tool for real-time visualization and monitoring of live data such as ML experiments. |
| 7 | [BertViz](https://awesomedataviz.com/tools/bertviz/) | 8,193 | 1 | Apache-2.0 | Visualize attention in Transformer language models such as BERT and GPT-2. |
| 8 | [TensorBoard](https://awesomedataviz.com/tools/tensorboard/) | 7,229 | 40 | Apache-2.0 | TensorFlow's visualization toolkit for metrics, model graphs, embeddings and more. |
| 9 | [Aim](https://awesomedataviz.com/tools/aim/) | 6,276 | 1 | Apache-2.0 | Experiment tracker with a UI to explore and compare ML runs and metrics. |
| 10 | [Embedding Atlas](https://awesomedataviz.com/tools/embedding-atlas/) | 4,965 | 134 | MIT | Interactive visualization of large embeddings with search, filtering and density views, by Apple. |
| 11 | [Yellowbrick](https://awesomedataviz.com/tools/yellowbrick/) | 4,407 | 0 | Apache-2.0 | Visual analysis and diagnostic tools for machine learning model selection with scikit-learn. |
| 12 | [LIT](https://awesomedataviz.com/tools/lit/) | 3,668 | 0 | Apache-2.0 | Learning Interpretability Tool: interactive visual analysis of ML model behavior, by Google PAIR. |
| 13 | [TensorWatch](https://awesomedataviz.com/tools/tensorwatch/) | 3,474 | 6 | MIT | Debugging and visualization tool for data science and machine learning |
| 14 | [dtreeviz](https://awesomedataviz.com/tools/dtreeviz/) | 3,159 | 18 | MIT | Decision tree visualization and model interpretation library for Python. |
| 15 | [Comet](https://awesomedataviz.com/tools/comet/) | 175 | 38 | MIT | An MLOps platform for tracking, visualizing, and debugging your machine learning workflows from training straight through to production. |

Source: https://awesomedataviz.com/categories/machine-learning/ · Data updated 2026-10-04 · Content: CC BY 4.0, Awesome Dataviz (https://awesomedataviz.com).