Download Latest Version v8.1.1 source code.zip (331.5 kB)
Email in envelope

Get an email when there's a new version of DifferentialEquations.jl

Home / v8.1.0
Name Modified Size InfoDownloads / Week
Parent folder
README.md 2026-08-23 4.1 kB
v8.1.0 source code.tar.gz 2026-08-23 326.9 kB
v8.1.0 source code.zip 2026-08-23 331.5 kB
Totals: 3 Items   662.4 kB 0

DifferentialEquations v8.1.0

Diff since v8.0.1

Merged pull requests:

  • QA: run_qa v1.6 form + ExplicitImports (#1152) (@ChrisRackauckas-Claude)
  • Release v8.0.2 (#1153) (@ChrisRackauckas)
  • Allow SciMLTesting v2 in compat (#1155) (@ChrisRackauckas-Claude)
  • Restrict SciMLTesting compat to v2.1 (#1156) (@ChrisRackauckas-Claude)
  • Enable rendered public API docs QA (#1157) (@ChrisRackauckas-Claude)
  • Update QA for SciMLTesting 2.4 (#1158) (@ChrisRackauckas-Claude)
  • Bump DifferentialEquations to 8.0.3 for release (#1162) (@ChrisRackauckas-Claude)
  • Drop older SciML major versions from compat (#1163) (@ChrisRackauckas-Claude)
  • QA: remove obsolete JET override and complete reexport allowlist (#1164) (@ChrisRackauckas-Claude)
  • Add PrecompileTools workload (#1167) (@ChrisRackauckas-Claude)
  • Bump versions for release (#1168) (@ChrisRackauckas)

Closed issues:

  • Move DEDataArray to its own DataArray package (#236)
  • DimensionMismatch error in Jacobian for stiff SecondOrderODE systems (#301)
  • inexacterror for SDEs with complex numbers (#443)
  • AbstractDiffEqOperator absent (#445)
  • Conversion of solution to DataFrame not consistent with documentation (#457)
  • Fractional Differential Equation not supported (#461)
  • Possible broadcast overload when using ArrayPartition with DEDataMatrix and ODEProblems with Control (#464)
  • MethodError on specifying iterative linear solver for Rosenbrock (#492)
  • New DEDataArray broadcast is order dependent (#532)
  • "ODEProblem: Warming:Instability detected. Aborting" when using Mutable Arrays with ManifoldProjection (#542)
  • SavedValues(Float64, Array{any}) leads to an Internal Error. (#592)
  • Problem with SplitODEProblem in more than one dimension (#596)
  • Multiple callbacks result in apparent solver hang (#600)
  • Numerical instability and PositiveDomain fails (#622)
  • PresetTimeCallback at t=0. triggers dt <= dtmin if f(u) = 0. (#625)
  • dt <= dtmin when using two discrete callbacks (#648)
  • Add checking for unsupported keyword arguments (#651)
  • Return type for automatic solver selection is not inferrable (#655)
  • using a VectorOfArray state with a stiff ODE solver can fail due to length calculations (#666)
  • Error with DDEProblem when u_0 is a LabelledArray with slices (#693)
  • equality == of Solutions stack overflow (#719)
  • SecondOrderODEProblem gives wrong result (#726)
  • Solving JumpProblem with VariableRateJump fails if saveat argument is added (#733)
  • SavingCallback error with keyword argument saveat=dt::Number (#754)
  • saveat and DiscreteCallback may save the some time point twice (#762)
  • Support indexing by name in save_idxs (#814)
  • MethodError: no method matching randexp(::TaskLocalRNG, ::Type{ComplexF64}) (#839)
  • PeriodicCallback not triggering with Rational numbers (#878)
  • A bug for calculating a jacobian (#888)
  • Inconsistent choice of default algorithm order when eltype is complex (#901)
  • Error : Automatic dt setting has the wrong sign. Exiting (#908)
  • deepcopy error on sparse analytic Jacobian EnsembleProblems (#927)
  • Auto algorithm selection does not work with Array of StaticArrays (#939)
  • Solution handling for ComponentArrays: how to extract a variable by name? (#957)
  • Broken type stability when using auto-switching ode solvers (#975)
  • Memory leak during auto-algorithm selection with EnsembleProblem (#1044)
  • Documentation Serialization Example No Longer Works (#1053)
  • Callback TerminateSteadyState() does not work with AutoVern7/9 and Rodas4/Rodas5P (#1058)
  • [CI Issue] Spell Check consistently failing on false positives in SVG file (#1120)
  • Uncaught exceptions complicate the API (#1146)
  • AB3 (and other multistep methods) fails to reinitialize after callback / discontinuity because integrator.derivative_discontinuity is reset too early (#1154)
  • Triage note: v8 re-export surface / Rodas5 are intentional (not regressions) (#1159)
  • With complex variables, ImplicitEM uses a complex datatype for the time variable (#1161)
Source: README.md, updated 2026-08-23