# neuronika/neuronika

Tensors and dynamic neural networks in pure Rust.

Repository: https://github.com/neuronika/neuronika
Canonical: https://ross.abutalabs.com/products/neuronika
Homepage: https://docs.rs/neuronika
Language: Rust
License: Apache-2.0
License Family: permissive
Topics: deeplearning, rust, autodifferentiation, neural-networks, rust-sci
Last push: 2022-10-10T07:02:35+00:00

## Health v2 (maintenance only)
Score: 32/100 (v2, computed 2026-09-02T17:46:02.011165+00:00)
- activity 0, release rhythm 35, longevity 100
- inputs: {"age_days": 2027, "days_push": 1423, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: no_releases
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 1086, forks 56 (observed 2026-08-28T04:03:32.110826+00:00)

## What it is
Neuronika is a machine learning framework written in pure Rust providing tensors and dynamic neural networks with reverse-mode automatic differentiation via a define-by-run API. It focuses on ease of use, fast prototyping, and performance without FFI, with optional CUDA and BLAS feature flags.

## Use cases
- build neural networks in pure Rust
- train models with reverse-mode autodifferentiation
- prototype deep learning models without Python
- do tensor math with a define-by-run API
- accelerate training with CUDA
- avoid FFI overhead in ML workloads

## When to choose
- you want a pure-Rust deep learning framework with no FFI
- you need dynamic, imperative define-by-run model definitions
- you prefer Rust's safety and performance for ML prototyping

## When to avoid
- you need a mature, actively maintained framework with a large ecosystem
- you require production-grade GPU support or extensive model zoo
- you need the latest deep learning features and community support

## Facets
- artifact type: library
- maturity: abandoned
- function: machine-learning, deep-learning, llm-training
- domain: machine-learning, deep-learning
- platform: rust, cross-platform
- tags: tensors, autodifferentiation, neural-networks, define-by-run, ndarray, pure-rust, gpu

## Member repositories
- neuronika/neuronika (main) score 32

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:03:32.110826+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 2026-08-30T06:49:36.714547+00:00, confidence not recorded.
  - readme: https://github.com/neuronika/neuronika (fetched 2026-08-28T04:03:32.110826+00:00, sha f8226e49b1f4)
  - homepage: https://docs.rs/neuronika (fetched 2026-08-29T12:52:19.387387+00:00, sha 70fd206dd9c9)
  - site_page: https://docs.rs/ (fetched 2026-08-29T12:52:19.396971+00:00, sha 9800f4a73bdd)
  - site_page: https://docs.rs/about (fetched 2026-08-29T12:52:19.398920+00:00, sha 29ab7986fbf5)
  - site_page: https://docs.rs/about/badges (fetched 2026-08-29T12:52:19.400665+00:00, sha 6588cac258af)
  - site_page: https://docs.rs/about/builds (fetched 2026-08-29T12:52:19.402320+00:00, sha 811231ca4680)
  - site_page: https://docs.rs/about/metadata (fetched 2026-08-29T12:52:19.403973+00:00, sha 0dcdccb2b2b2)
  - site_page: https://docs.rs/about/redirections (fetched 2026-08-29T12:52:19.405617+00:00, sha 1b1efd93105f)
  - site_page: https://docs.rs/about/download (fetched 2026-08-29T12:52:19.407509+00:00, sha 6793b0ab6541)
  - site_page: https://docs.rs/about/rustdoc-json (fetched 2026-08-29T12:52:19.409269+00:00, sha 5b94b2d8e2ff)
- Data as of 2026-08-30T08:39:29.467469+00:00.