# JuliaDiff/ForwardDiff.jl

Forward Mode Automatic Differentiation for Julia

Repository: https://github.com/JuliaDiff/ForwardDiff.jl
Canonical: https://ross.abutalabs.com/products/forwarddiffjl
Language: Julia
License: NOASSERTION
License Family: other
Topics: julia, calculus, automatic-differentiation
Last push: 2026-09-01T14:03:57+00:00

## Health v2 (maintenance only)
Score: 99/100 (v2, computed 2026-09-02T17:46:02.011165+00:00)
- activity 100, release rhythm 96, longevity 100
- inputs: {"age_days": 4890, "days_push": 1, "days_rel": 27, "gap_med": 20.0, "n_releases_24m": 19}
- 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 1004, forks 160 (observed 2026-09-03T02:15:06.312742+00:00)

## What it is
ForwardDiff.jl is a Julia library implementing forward mode automatic differentiation for computing derivatives, gradients, Jacobians, Hessians, and higher-order derivatives of native Julia functions. It uses dual numbers to deliver exact derivatives that generally outperform finite-differencing in both speed and accuracy.

## Use cases
- compute gradients of Julia functions for optimization
- calculate Jacobians of vector-valued functions
- compute Hessians for second-order optimization
- take derivatives of scalar functions exactly
- replace finite-difference approximations with exact AD
- differentiate arbitrary callable objects in Julia

## When to choose
- you need exact derivatives of Julia code without manual derivation
- your functions map scalars to vectors or have moderate input dimensionality
- you want faster and more accurate results than finite differencing
- you are doing scientific computing or optimization in Julia

## When to avoid
- your function has very large input dimensions where reverse-mode AD is more efficient
- you are not working in the Julia language
- your code uses operations that AD cannot differentiate through

## Facets
- artifact type: library
- maturity: stable
- function: math
- domain: machine-learning, data-science
- platform: jvm-scripting
- tags: julia, automatic-differentiation, forward-mode-ad, calculus, dual-numbers, gradients, jacobians, hessians, scientific-computing, optimization, algorithms

## Member repositories
- JuliaDiff/ForwardDiff.jl (main) score 99

## Provenance
- Observed fields: from GitHub, fetched 2026-09-03T02:15:06.312742+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 2026-08-30T07:13:48.319811+00:00, confidence not recorded.
  - readme: https://github.com/JuliaDiff/ForwardDiff.jl (fetched 2026-09-03T02:15:06.312742+00:00, sha 2600a10d06b0)
- Data as of 2026-08-30T08:39:29.467469+00:00.