# jolibrain/deepdetect

Deep Learning Server and CLI for Torch and TensorRT

Repository: https://github.com/jolibrain/deepdetect
Canonical: https://ross.abutalabs.com/products/deepdetect
Homepage: https://www.deepdetect.com/
Language: C++
License: NOASSERTION
License Family: other
Topics: deep-learning, machine-learning, caffe, xgboost, rest-api, tsne, object-detection, image-segmentation, image-classification, neural-nets, gpu, ncnn, tensorrt, time-series, pytorch, tensorrt-conversion, tensorrt-inference, image-search
Last push: 2026-08-26T15:08:35+00:00

## Health v2 (maintenance only)
Score: 95/100 (v2, computed 2026-09-02T17:46:02.011165+00:00)
- activity 99, release rhythm 87, longevity 100
- inputs: {"age_days": 4121, "days_push": 7, "days_rel": 7, "gap_med": 88.5, "n_releases_24m": 7}
- flags: no_license
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 2551, forks 546 (observed 2026-08-28T04:07:00.543690+00:00)

## What it is
DeepDetect is an open-source deep learning runtime, CLI, and REST server written in C++ for training and inference across images, text, tabular data, and time series. It supports PyTorch, TensorRT, and other backends, with a web platform UI and Docker-based deployment for CPU and GPU.

## Use cases
- serve object detection models over a REST API
- train and run image segmentation models
- deploy deep learning inference on GPU with TensorRT
- classify images and text via a single API
- run predictions on tabular CSV and time-series data
- manage model training jobs and repositories from a CLI
- self-host a deep learning server with Docker

## When to choose
- you need a production-ready REST server for model training and inference
- you want one API surface for detection, segmentation, classification, and tabular tasks
- you need GPU-accelerated serving with TensorRT or PyTorch on Linux
- you prefer a self-hosted, C++-fast alternative to managed ML serving

## When to avoid
- you need a pure Python training framework with full ecosystem flexibility
- you only want lightweight experimentation in notebooks without a server
- you require Windows-native GPU serving or non-Linux production targets
- you need the latest cutting-edge model architectures immediately after release

## Facets
- artifact type: service
- maturity: active
- function: machine-learning, deep-learning, llm-inference, image-processing, http-server, api-framework, cli, gpu-computing
- domain: deep-learning, machine-learning, computer-vision, apis, developer-tools
- platform: python, cpp, cross-platform, self-hosted
- tags: rest-api, object-detection, image-segmentation, image-classification, tensorrt, pytorch, model-serving, training, inference, time-series, ocr, natural-language-processing, linux, docker, gpu

## Member repositories
- jolibrain/deepdetect (main) score 95

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:07:00.543690+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 2026-08-30T02:23:55.615368+00:00, confidence not recorded.
  - readme: https://github.com/jolibrain/deepdetect (fetched 2026-08-28T04:07:00.543690+00:00, sha 54793f227295)
  - homepage: https://www.deepdetect.com/ (fetched 2026-08-29T10:06:41.030554+00:00, sha 5b86631034fb)
  - site_page: https://www.deepdetect.com/quickstart-platform (fetched 2026-08-29T10:06:41.039571+00:00, sha d2cb5dcd4b58)
  - site_page: https://www.deepdetect.com/quickstart-server (fetched 2026-08-29T10:06:41.041479+00:00, sha 61a606209f39)
- Data as of 2026-08-30T08:39:29.467469+00:00.