Top AI Repos — open-source AI, indexed and scored
Top AI Repos tracks AI repositories on GitHub and answers two different questions about each one: is it moving right now, and would you bet a product on it.
Top AI Repos tracks AI repositories on GitHub and answers two different questions about each one: is it moving right now, and would you bet a product on it.
Tangle is a web app that allows the users to build and run Machine Learning pipelines without having to set up development environment.
| Date | Stars |
|---|---|
| 2026-07-31 | 263 |
| 2026-08-06 | 265 |
| 2026-08-07 | 266 |
| 2026-08-08 | 266 |
| 2026-08-10 | 266 |
| 2026-08-11 | 266 |
| 2026-08-12 | 266 |
| 2026-08-14 | 266 |
| 2026-08-15 | 266 |
| 2026-08-18 | 266 |
| 2026-08-20 | 266 |
| 2026-08-22 | 266 |
| 2026-08-28 | 267 |
| 2026-08-29 | 267 |
| 2026-09-02 | 395 |
| 2026-09-03 | 458 |
| 2026-09-04 | 474 |
| 2026-09-05 | 481 |
| 2026-09-06 | 486 |
| 2026-09-07 | 497 |
| 2026-09-08 | 498 |
| 2026-09-09 | 499 |
| 2026-09-10 | 501 |
| 2026-09-11 | 505 |
| 2026-09-15 | 505 |
| 2026-09-17 | 506 |
| 2026-09-19 | 507 |
| 2026-09-20 | 508 |
Today
+1 stars today
This week
+3 stars this week
This month
+242 stars this month
Momentum
7.0
growth rate 0.59%/day
# Tangle The Tangle system helps users create and run ML experiments and production pipelines. Any batch workflow that has beginning and end can be orchestrated via a pipeline. To learn more about Tangle, visit <https://tangleml.com> [](https://tangleml-tangle.hf.space/#/quick-start) ## Try Tangle Tangle’s architecture allows it to run anywhere. We’re expanding the list of installation recipes. At this moment Tangle can be used locally (using Docker/Podman launcher) or at HuggingFace (using HuggingFace Job launcher). ### Try at HuggingFace The easiest way to try Tangle is via HuggingFace. Go to [TangleML/tangle](https://tangleml-tangle.hf.space/#/quick-start) at HuggingFace and start building. Creating a pipeline does not require any registration, but to run your pipeline, you’ll need a HuggingFace account with [Pro](https://huggingface.co/pro) subscription ($9/month). ### Build a pipeline Start with the sample XGBoost training pipeline or build a new one from scratch: * Drag components onto canvas * Connect components (outputs to inputs) * Configure task arguments * Submit pipeline for execution (requires login) * Monitor the pipeline run in real-time ### HuggingFace x Tangle integration The Tangle team has deployed Tangle to HuggingFace Spaces as a multi-tenant service, which uses HF infrastructure for storage, compute, and authentication. The shared multi-tenant instance is live at <https://tangleml-tangle.hf.space> (or <https://huggingface.co/spaces/TangleML/tangle>). Tangle’s multi-tenant architecture maintains a central tenant database (storing user IDs, access tokens, and orchestrator configs) plus individual per-tenant SQLite databases in the main TangleML/tangle HuggingFace Space persistent storage. Each user’s pipelines run via that user’s HuggingFace Jobs, and the execution logs and output artifacts are stored in the user's own private HuggingFace Dataset repo (user/tangle_data), with clickable links in the UI to both artifacts and HuggingFace Jobs. You can also deploy your own single-tenant instance. To do this, [duplicate](https://huggingface.co/spaces/TangleML/tangle?duplicate=true) the Tangle space to your HF account and provide a [HuggingFace token](https://huggingface.co/docs/hub/en/security-tokens). These single-tenant Tangle deployments store their database in your own HF Space persistent storage, giving you complete control and data isolation. If you duplicate Tangle to an organization, you’ll get a single-tenant multi-user Tangle deployment for your team, where multiple team members can see each other’s pipeline runs and enjoy org-wide cache. You can also deploy Tangle to other environments (local or cloud). ## Installation ### Try on local machine 1. Install [Docker](https://www.docker.com/get-started/) and [uv](https://docs.astral.sh/uv/getting-started/installation/). 2. Download the app code (needs to be done once): ```shell git clone https://github.com/TangleML/tangle.git tangle/backend --branch stable git clone https://github.com/TangleML/tangle-ui.git tangle/ui_build --branch stable_local_build ``` 3. Start the app: Linux and Mac OS: ```shell cd tangle && backend/start_local.sh ``` Windows: ```bat cd tangle && backend\start_local.cmd ``` 4. Once the "start_local: Starting the orchestrator" message appears in the terminal, open the [http://localhost:8000](http://localhost:8000) URL in a Web browser and start use the app. Click the "New Pipeline" button at the top to start building a new pipeline. ### Try in Google Cloud Shell (free) [Google Cloud Shell](https://cloud.google.com/shell/) is free (50 hours per week) and needs a Google Cloud account. 1. Open [Google Cloud Shell](https://shell.cloud.google.com/?show=terminal) in a Web browser 2. Download the app code (needs to be done once): ```shell git clone https://github.com/TangleML/tangle.git tangle/backend --branch stab
Excerpt of 9,262 characters
Read on GitHub
Alexey Volkov · @shopify
295
46
41
24
1
Mark Zhang · United States
1
1
1
Zhanghao Wu · Sky Computing Lab, UC Berkeley · United States
1
Would you bet a product on this? Bounded 0–100 and slow moving.
matched fp:75778c70165f93ad, llm:Description: 'Tangle is a web app that allows the users to build and run Machine Learning pipelines without having to set up development environment.' Topics: artificial-intelligence, cloud-pipelines, data-processing, machine-learning, pipelines
matched fp:75778c70165f93ad, llm:Description: 'Tangle is a web app that allows the users to build and run Machine Learning pipelines without having to set up development environment.' Topics: artificial-intelligence, cloud-pipelines, data-processing, machine-learning, pipelines
matched fp:75778c70165f93ad, llm:Description: 'Tangle is a web app that allows the users to build and run Machine Learning pipelines without having to set up development environment.' Topics: artificial-intelligence, cloud-pipelines, data-processing, machine-learning, pipelines