GLPK wrapper module for Julia
98 Stars
Updated Last
1 Year Ago
Started In
February 2013


Build Status codecov

GLPK.jl is a wrapper for the GNU Linear Programming Kit library.

The wrapper has two components:


This wrapper is maintained by the JuMP community and is not an GNU project.


GLPK.jl is licensed under the GPL v3 license.


Install GLPK using Pkg.add:

import Pkg

In addition to installing the GLPK.jl package, this will also download and install the GLPK binaries. You do not need to install GLPK separately.

To use a custom binary, read the Custom solver binaries section of the JuMP documentation.

Use with JuMP

To use GLPK with JuMP, use GLPK.Optimizer:

using JuMP, GLPK
model = Model(GLPK.Optimizer)
set_attribute(model, "tm_lim", 60 * 1_000)
set_attribute(model, "msg_lev", GLPK.GLP_MSG_OFF)

If the model is primal or dual infeasible, GLPK will attempt to find a certificate of infeasibility. This can be expensive, particularly if you do not intend to use the certificate. If this is the case, use:

model = Model(() -> GLPK.Optimizer(; want_infeasibility_certificates = false))

MathOptInterface API

The GLPK optimizer supports the following constraints and attributes.

List of supported objective functions:

List of supported variable types:

List of supported constraint types:

List of supported model attributes:


Options for GLPK are comprehensively documented in the PDF documentation, but they are hard to find.

  • Options when solving a linear program are defined in Section 2.8.1
  • Options when solving a mixed-integer program are defined in Section 2.10.5

However, the following options are likely to be the most useful:

Parameter Example Explanation
msg_lev GLPK.GLP_MSG_ALL Message level for terminal output
presolve GLPK.GLP_ON Turn presolve on or off
tol_int 1e-5 Absolute tolerance for integer feasibility
tol_obj 1e-7 Relative objective tolerance for mixed-integer programs


Here is an example using GLPK's solver-specific callbacks.

using JuMP, GLPK, Test

model = Model(GLPK.Optimizer)
@variable(model, 0 <= x <= 2.5, Int)
@variable(model, 0 <= y <= 2.5, Int)
@objective(model, Max, y)
reasons = UInt8[]
function my_callback_function(cb_data)
    reason = GLPK.glp_ios_reason(cb_data.tree)
    push!(reasons, reason)
    if reason != GLPK.GLP_IROWGEN
    x_val = callback_value(cb_data, x)
    y_val = callback_value(cb_data, y)
    if y_val - x_val > 1 + 1e-6
        con = @build_constraint(y - x <= 1)
        MOI.submit(model, MOI.LazyConstraint(cb_data), con)
    elseif y_val + x_val > 3 + 1e-6
        con = @build_constraint(y - x <= 1)
        MOI.submit(model, MOI.LazyConstraint(cb_data), con)
MOI.set(model, GLPK.CallbackFunction(), my_callback_function)
@test termination_status(model) == MOI.OPTIMAL
@test primal_status(model) == MOI.FEASIBLE_POINT
@test value(x) == 1
@test value(y) == 2
@show reasons


The C API can be accessed via GLPK.glp_XXX functions, where the names and arguments are identical to the C API. See the /tests folder for inspiration.