#8484th of 105 in ML frameworks and libraries
tensorboard
Web dashboards for inspecting TensorFlow training runs, graphs and embeddings
- Stars
- 7.2k
- License
- Apache-2.0
- Last commit
- Oct 2026
- Last release
- Jun 2026
- Language
- TypeScript
Missed the latest GitHub check; these facts may be older.
Overview
TensorBoard reads tfevents files from a log directory and serves a local web UI, by default on port 6006, for inspecting training runs. It has scalar, histogram, distribution, image, audio, text and time series dashboards, plus a graph explorer and an embedding projector. It runs offline and supports third-party plugins published on PyPI.
Who it is for: ML engineers tracking TensorFlow training runs locally or on a private network
Strengths
- Runs fully offline against a local log directory with one `tensorboard --logdir` command
- Compares runs by recursively loading every subdirectory that contains tfevents data
- Plugin system documented in ADDING_A_PLUGIN, with example plugins and PyPI distribution
- Covers scalars, histograms, images, audio, text, model graphs and an embedding projector
Weaknesses
- Reads one events file at a time; multiple writers need the experimental --reload_multifile flag
- Reservoir-samples data to fit in RAM, so plots may not show every point
- No native scatterplots or custom plots; workaround is writing matplotlib images
- README names only Chrome and Firefox as supported browsers
What it needs
- no GPU
- Docker
- Needs TensorFlow summary event files (tfevents), Bazel (only to build from source)
- Models: TensorFlow
- port 6006
Also in ML frameworks and libraries
See all 105| Rank | Project | Score |
|---|---|---|
| 1 | transformersPython library defining open models for text, vision, audio and multimodal tasks | 94 out of 100 |
| 2 | pytorchPython tensor library with GPU support and tape-based autograd for deep learning | 94 out of 100 |
| 3 | ultralyticsPython library and CLI for training and running YOLO vision models | 93 out of 100 |
| #4 | pytorch-image-modelsPyTorch library of image models, pretrained weights and reference training scripts | 92 out of 100 |
| #5 | supervisionPython toolkit for loading, annotating and counting computer vision detections | 91 out of 100 |