# determined-ai/determined

Determined is an open-source machine learning platform that simplifies distributed training, hyperparameter tuning, experiment tracking, and resource management. Works with PyTorch and TensorFlow.

Repository: https://github.com/determined-ai/determined
Canonical: https://ross.abutalabs.com/products/determined
Homepage: https://determined.ai
Language: Go
License: Apache-2.0
License Family: permissive
Topics: deep-learning, machine-learning, ml-platform, ml-infrastructure, hyperparameter-optimization, hyperparameter-search, distributed-training, pytorch, tensorflow, hyperparameter-tuning, kubernetes, data-science, mlops, keras
Last push: 2025-03-20T19:09:46+00:00

## Health v2 (maintenance only)
Score: 39/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 12, release rhythm 40, longevity 100
- inputs: {"age_days": 2339, "days_push": 531, "days_rel": 532, "gap_med": 13, "n_releases_24m": 6}
- flags: none
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 3236, forks 373 (observed 2026-08-28T04:07:50.319291+00:00)

## What it is
Determined is an open-source deep learning platform that combines distributed training, hyperparameter tuning, experiment tracking, and GPU resource management in one system. It works with PyTorch and TensorFlow via a Python library, CLI, and Web UI, and deploys on-prem, on AWS/GCP, Kubernetes, or Slurm.

## Use cases
- run distributed deep learning training on multiple GPUs
- tune hyperparameters automatically for PyTorch or TensorFlow models
- track and compare ML experiments with reproducibility
- manage GPU clusters and cut cloud training costs
- deploy a training cluster on AWS, GCP, or Kubernetes
- visualize loss curves and hyperparameter search results in a web UI

## When to choose
- you need an all-in-one platform for training, tuning, and tracking deep learning experiments
- you want to scale PyTorch/TensorFlow training across multiple GPUs or nodes
- you need efficient GPU scheduling and resource management for a team
- you want reproducible experiment configs via YAML

## When to avoid
- you only need lightweight experiment tracking without cluster management
- your workflow is centered on a single GPU with no distributed training
- you prefer assembling your own stack from separate tools like MLflow and Ray

## Facets
- artifact type: application
- maturity: active
- function: machine-learning, llm-training, gpu-computing, scheduling, monitoring, cli, developer-tools
- domain: machine-learning, deep-learning, data-science, cloud-computing, gpu-computing, developer-tools
- platform: python, go, cloud, self-hosted, cli, cross-platform
- tags: mlops, hyperparameter-tuning, distributed-training, experiment-tracking, pytorch, tensorflow, resource-management, ml-platform, kubernetes, docker, web-server

## Member repositories
- determined-ai/determined (main) score 39

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:07:50.319291+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-30T07:24:35.163499+00:00, confidence not recorded.
  - readme: https://github.com/determined-ai/determined (fetched 2026-08-28T04:07:50.319291+00:00, sha d560e112e74e)
- Data as of 2026-08-30T08:39:29.467469+00:00.