# clab/dynet

DyNet: The Dynamic Neural Network Toolkit

Repository: https://github.com/clab/dynet
Canonical: https://ross.abutalabs.com/products/dynet
Language: C++
License: Apache-2.0
License Family: permissive
Last push: 2023-12-01T17:10:01+00:00

## Health v2 (maintenance only)
Score: 23/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 0, release rhythm 8, longevity 100
- inputs: {"age_days": 4224, "days_push": 1006, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: none
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 3436, forks 701 (observed 2026-08-28T04:08:04.964471+00:00)

## What it is
DyNet is a C++ neural network toolkit with Python bindings, designed for efficient CPU/GPU training of networks with dynamic per-instance structure. It features auto-batching and has powered NLP systems for parsing, machine translation, and morphological inflection.

## Use cases
- build neural networks with dynamic computation graphs
- train NLP models like parsers and machine translation systems
- run deep learning efficiently on CPU or GPU
- auto-batch many small training examples for speed
- prototype research models in Python or C++

## When to choose
- your model's graph structure changes per training instance
- you need fast auto-batched training for NLP research
- you want a lightweight C++ deep learning library with Python bindings

## When to avoid
- you need a large ecosystem of pretrained models and tools
- you want mainstream community support like PyTorch or TensorFlow
- the project's low recent activity is a concern for long-term use

## Facets
- artifact type: library
- maturity: maintenance
- function: machine-learning, deep-learning, nlp
- domain: machine-learning, deep-learning
- platform: cpp, python, windows
- tags: dynamic-computation-graph, auto-batching, neural-networks, tensor-library, natural-language-processing, linux, macos, gpu

## Member repositories
- clab/dynet (main) score 23

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:08:04.964471+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 2026-08-29T18:37:41.823212+00:00, confidence not recorded.
  - readme: https://github.com/clab/dynet (fetched 2026-08-28T04:08:04.964471+00:00, sha 377588c6eda6)
- Data as of 2026-08-30T08:39:29.467469+00:00.