add cuopt direct solver #3620

Iroy30 · 2025-06-04T18:31:12Z

Fixes #3626

Summary/Motivation:

Add cuOpt math optimization (includes LP and MILP) solver backend to Pyomo so users can solve pyomo models with cuOpt

Changes proposed in this PR:

Add cuopt_direct solver plugin

Legal Acknowledgement

By contributing to this software project, I have read the contribution guide and agree to the following terms and conditions for my contribution:

I agree my contributions are submitted under the BSD license.
I represent I am authorized to make the contributions and grant the license. If my employer has rights to intellectual property that includes these contributions, I represent that I have received permission to make contributions and grant the required license on behalf of that employer.

codecov · 2025-06-17T05:20:43Z

Codecov Report

❌ Patch coverage is 91.04478% with 24 lines in your changes missing coverage. Please review.
✅ Project coverage is 89.45%. Comparing base (e0fcc81) to head (ed0d4bb).

Files with missing lines	Patch %	Lines
pyomo/solvers/plugins/solvers/cuopt_direct.py	90.97%	24 Missing ⚠️

Additional details and impacted files

@@           Coverage Diff            @@
##             main    #3620    +/-   ##
========================================
  Coverage   89.45%   89.45%            
========================================
  Files         905      906     +1     
  Lines      105467   105735   +268     
========================================
+ Hits        94341    94586   +245     
- Misses      11126    11149    +23

Flag	Coverage Δ
builders	`29.12% <20.14%> (-0.03%)`	⬇️
default	`86.08% <91.04%> (?)`
expensive	`35.73% <20.52%> (?)`
linux	`86.59% <19.40%> (-2.62%)`	⬇️
linux_other	`86.59% <19.40%> (-0.18%)`	⬇️
osx	`82.75% <19.40%> (-0.17%)`	⬇️
win	`84.83% <19.40%> (-0.17%)`	⬇️
win_other	`84.83% <19.40%> (-0.17%)`	⬇️

Flags with carried forward coverage won't be shown. Click here to find out more.

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

🚀 New features to boost your workflow:

❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.

mrmundt

We are working on making cuopt available in our testing infrastructure; can you please add tests to this PR?

mrmundt · 2025-07-01T15:57:07Z

@Iroy30 - We've been able to make cuopt available on our internal testing machines. Can you please add tests to this PR?

Iroy30 · 2025-07-02T12:38:58Z

@mrmundt Thanks! We have added tests by enabling testing cuopt with LP and MILP capabilities in tests/solvers.py. Let us know if:

There is any other testing that needs to be enabled.
For testing cuOpt backend in CI, it needs a GPU, is that something pyomo team intends to add or will it be disabled in github CI?

The following is the testing output I get relevant to cuOpt

Testing scenario generation
------------------------------
LP_block, cuopt, python
   ok: 
LP_block_kernel, cuopt, python
   ok: 
LP_compiled, cuopt, python
   ok: 
LP_compiled_dense_kernel, cuopt, python
   ok: 
LP_compiled_sparse_kernel, cuopt, python
   ok: 
LP_constant_objective1, cuopt, python
   ok: 
LP_constant_objective1_kernel, cuopt, python
   ok: 
LP_constant_objective2, cuopt, python
   ok: 
LP_constant_objective2_kernel, cuopt, python
   ok: 
LP_duals_maximize, cuopt, python
   ok: 
LP_duals_maximize_kernel, cuopt, python
   ok: 
LP_duals_minimize, cuopt, python
   ok: 
LP_duals_minimize_kernel, cuopt, python
   ok: 
LP_inactive_index, cuopt, python
   ok: 
LP_inactive_index_kernel, cuopt, python
   ok: 
LP_infeasible1, cuopt, python
   ok: 
LP_infeasible1_kernel, cuopt, python
   ok: 
LP_infeasible2, cuopt, python
   ok: 
LP_infeasible2_kernel, cuopt, python
   ok: 
LP_piecewise, cuopt, python
   ok: 
LP_piecewise_nosuffixes, cuopt, python
   ok: 
LP_simple, cuopt, python
   ok: 
LP_simple_kernel, cuopt, python
   ok: 
LP_trivial_constraints, cuopt, python
   ok: 
LP_trivial_constraints_kernel, cuopt, python
   ok: 
LP_unbounded, cuopt, python
   ok: 
LP_unbounded_kernel, cuopt, python
   ok: 
LP_unused_vars, cuopt, python
   ok: 
LP_unused_vars_kernel, cuopt, python
   ok: 
MILP_discrete_var_bounds, cuopt, python
   ok: 
MILP_discrete_var_bounds_kernel, cuopt, python
   ok: 
MILP_infeasible1, cuopt, python
   ok: 
MILP_infeasible1_kernel, cuopt, python
   ok: 
MILP_simple, cuopt, python
   ok: 
MILP_simple_kernel, cuopt, python
   ok: 
MILP_unbounded, cuopt, python
   ok: 
MILP_unbounded_kernel, cuopt, python
   ok: 
MILP_unused_vars, cuopt, python
   ok: 
MILP_unused_vars_kernel, cuopt, python
   ok:

pyomo/solvers/plugins/solvers/cuopt_direct.py

mrmundt · 2025-07-02T13:35:58Z

pyomo/solvers/plugins/solvers/cuopt_direct.py

+        t0 = time.time()
+        self.solution = cuopt.linear_programming.solver.Solve(self._solver_model)
+        t1 = time.time()


This is fine, but just so you're aware, we have this lovely little utility called TicTocTimer that you may want to consider using: https://pyomo.readthedocs.io/en/latest/api/pyomo.common.timing.TicTocTimer.html

whart222 · 2025-07-08T18:49:09Z

@mrmundt What do you think about including this solver interface in pyomo.contrib.solvers? Would it make sense to pull-in new solver interfaces there, since that's where the new solver API is evolving?

mrmundt · 2025-07-08T18:53:06Z

@mrmundt What do you think about including this solver interface in pyomo.contrib.solvers? Would it make sense to pull-in new solver interfaces there, since that's where the new solver API is evolving?

@whart222 - I am evenly split. Because we are still messing with what the new solver interfaces are going to actually do / how they will handle input and present output, I don't know if we want to put "new" solvers there or just "well-established" ones that we can robustly test / really know what they are supposed to do and return.

mrmundt · 2025-07-08T18:57:44Z

@Iroy30 - I forgot to post this last week, but all of the failures are of the variety:

self = <pyomo.solvers.tests.checks.test_no_solution_behavior.Test_LP_unbounded_kernel testMethod=test_cuopt_python>

    def return_test(self):
>       return failed_solve_test(self)

/tester/python/python39/pyomo/pyomo/solvers/tests/checks/test_no_solution_behavior.py:92: 
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ 
/tester/python/python39/pyomo/pyomo/solvers/tests/checks/test_no_solution_behavior.py:61: in failed_solve_test
    opt, results = model_class.solve(
/tester/python/python39/pyomo/pyomo/solvers/tests/models/base.py:123: in solve
    results = opt.solve(
/tester/python/python39/pyomo/pyomo/opt/base/solvers.py:99: in solve
    self._solver_error('solve')
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ 

self = <pyomo.opt.base.solvers.UnknownSolver object at 0x7f86ad9ab040>
method_name = 'solve'

        def _solver_error(self, method_name):
>           raise RuntimeError(
                """Attempting to use an unavailable solver.
    
    The SolverFactory was unable to create the solver "%s"
    and returned an UnknownSolver object.  This error is raised at the point
    where the UnknownSolver object was used as if it were valid (by calling
    method "%s").
    
    The original solver was created with the following parameters:
    \t"""
                % (self.type, method_name)
                + "\n\t".join("%s: %s" % i for i in sorted(self._kwds.items()))
                + "\n\t_args: %s" % (self._args,)
                + "\n\toptions: %s" % (self.options,)
            )
E           RuntimeError: Attempting to use an unavailable solver.
E           
E           The SolverFactory was unable to create the solver "cuopt"
E           and returned an UnknownSolver object.  This error is raised at the point
E           where the UnknownSolver object was used as if it were valid (by calling
E           method "solve").
E           
E           The original solver was created with the following parameters:
E           	executable: cuopt
E           	solver_io: python
E           	type: cuopt
E           	_args: ()
E           	options: {}

/tester/python/python39/pyomo/pyomo/opt/base/solvers.py:116: RuntimeError

mrmundt · 2025-07-08T19:11:11Z

@Iroy30 - Two more things:

Please run black -S -C on your files
Something in our testing environment isn't loading correctly. I'm looking into it.

Co-authored-by: Miranda Mundt <[email protected]>

jsiirola · 2025-11-14T22:13:04Z

OK. We are getting close. cuopt finally appears to be running on Jenkins. We are seeing 9 failures:

FAILED pyomo/pyomo/solvers/tests/checks/test_no_solution_behavior.py::Test_MILP_unbounded_kernel::test_cuopt_python - AssertionError: <TerminationCondition.infeasible: 'infeasible'> not found in (<TerminationCondition.unbounded: 'unbounded'>, <TerminationCondition.infeasibleOrUnbounded: 'infeasibleOrUnbounded'>)
FAILED pyomo/pyomo/solvers/tests/checks/test_writers.py::Test_LP_duals_maximize::test_cuopt_python_nonsymbolic_labels - AssertionError: Solution mismatch for plugin test_cuopt_python_nonsymbolic_labels, python interface and problem type LP_duals_maximize
Difference in solution for x[1].rc:
	Baseline - -1.0
	Current - 1.0
FAILED pyomo/pyomo/solvers/tests/checks/test_writers.py::Test_LP_duals_maximize::test_cuopt_python_symbolic_labels - AssertionError: Solution mismatch for plugin test_cuopt_python_symbolic_labels, python interface and problem type LP_duals_maximize
Difference in solution for x[1].rc:
	Baseline - -1.0
	Current - 1.0
FAILED pyomo/pyomo/solvers/tests/checks/test_writers.py::Test_LP_duals_maximize_kernel::test_cuopt_python_nonsymbolic_labels - AssertionError: Solution mismatch for plugin test_cuopt_python_nonsymbolic_labels, python interface and problem type LP_duals_maximize
Difference in solution for x[1].rc:
	Baseline - -1.0
	Current - 1.0
FAILED pyomo/pyomo/solvers/tests/checks/test_writers.py::Test_LP_duals_maximize_kernel::test_cuopt_python_symbolic_labels - AssertionError: Solution mismatch for plugin test_cuopt_python_symbolic_labels, python interface and problem type LP_duals_maximize
Difference in solution for x[1].rc:
	Baseline - -1.0
	Current - 1.0
FAILED pyomo/pyomo/solvers/tests/checks/test_writers.py::Test_MILP_unbounded::test_cuopt_python_nonsymbolic_labels - AssertionError: <TerminationCondition.infeasible: 'infeasible'> not found in (<TerminationCondition.unbounded: 'unbounded'>, <TerminationCondition.infeasibleOrUnbounded: 'infeasibleOrUnbounded'>)
FAILED pyomo/pyomo/solvers/tests/checks/test_writers.py::Test_MILP_unbounded::test_cuopt_python_symbolic_labels - AssertionError: <TerminationCondition.infeasible: 'infeasible'> not found in (<TerminationCondition.unbounded: 'unbounded'>, <TerminationCondition.infeasibleOrUnbounded: 'infeasibleOrUnbounded'>)
FAILED pyomo/pyomo/solvers/tests/checks/test_writers.py::Test_MILP_unbounded_kernel::test_cuopt_python_nonsymbolic_labels - AssertionError: <TerminationCondition.infeasible: 'infeasible'> not found in (<TerminationCondition.unbounded: 'unbounded'>, <TerminationCondition.infeasibleOrUnbounded: 'infeasibleOrUnbounded'>)
FAILED pyomo/pyomo/solvers/tests/checks/test_writers.py::Test_MILP_unbounded_kernel::test_cuopt_python_symbolic_labels - AssertionError: <TerminationCondition.infeasible: 'infeasible'> not found in (<TerminationCondition.unbounded: 'unbounded'>, <TerminationCondition.infeasibleOrUnbounded: 'infeasibleOrUnbounded'>)

Iroy30 · 2025-11-17T06:32:08Z

OK. We are getting close. cuopt finally appears to be running on Jenkins. We are seeing 9 failures:

Fixed by deleting instance level version , now the tests are picked up correctly as expected failures (cuOpt has these known failures)

Iroy30 · 2025-11-19T03:05:36Z

Great! Looks like the CI passes except codecov/patch , how do we reconcile this, any suggestions?

jdwit · 2025-11-24T09:42:17Z

Great! Looks like the CI passes except codecov/patch , how do we reconcile this, any suggestions?

The codecov/patch check is failing because cuopt_direct.py has some uncovered lines (hitting ~77% coverage and ~90% is required). The fix is to add some more tests to test_cuopt_direct.py. Let me know if I can help out with this. Btw nice work on this - I'm curious how some of my simulations in Pyomo perform using cuopt.

Iroy30 · 2026-01-14T19:13:11Z

Hi all, does this look good for a final review and merge?

mrmundt · 2026-01-23T17:03:55Z

pyomo/solvers/plugins/solvers/cuopt_direct.py

+import logging
+import re
+import sys
+
+from pyomo.common.collections import ComponentSet, ComponentMap, Bunch
+from pyomo.common.dependencies import attempt_import
+from pyomo.common.dependencies import numpy as np
+from pyomo.core.base import Suffix, Var, Constraint, SOSConstraint, Objective
+from pyomo.common.errors import ApplicationError
+from pyomo.common.tempfiles import TempfileManager
+from pyomo.common.tee import capture_output
+from pyomo.core.expr.numvalue import is_fixed
+from pyomo.core.expr.numvalue import value
+from pyomo.core.staleflag import StaleFlagManager
+from pyomo.repn import generate_standard_repn
+from pyomo.repn.linear import LinearRepnVisitor
+from pyomo.solvers.plugins.solvers.direct_solver import DirectSolver
+from pyomo.solvers.plugins.solvers.direct_or_persistent_solver import (
+    DirectOrPersistentSolver,
+)
+from pyomo.common.enums import minimize, maximize
+from pyomo.opt.results.results_ import SolverResults
+from pyomo.opt.results.solution import Solution, SolutionStatus
+from pyomo.opt.results.solver import TerminationCondition, SolverStatus
+from pyomo.opt.base import SolverFactory
+import time


There are quite a few unused imports here (flagged by my IDE).

Suggested change

import logging

import re

import sys

from pyomo.common.collections import ComponentSet, ComponentMap, Bunch

from pyomo.common.dependencies import attempt_import

from pyomo.common.dependencies import numpy as np

from pyomo.core.base import Suffix, Var, Constraint, SOSConstraint, Objective

from pyomo.common.errors import ApplicationError

from pyomo.common.tempfiles import TempfileManager

from pyomo.common.tee import capture_output

from pyomo.core.expr.numvalue import is_fixed

from pyomo.core.expr.numvalue import value

from pyomo.core.staleflag import StaleFlagManager

from pyomo.repn import generate_standard_repn

from pyomo.repn.linear import LinearRepnVisitor

from pyomo.solvers.plugins.solvers.direct_solver import DirectSolver

from pyomo.solvers.plugins.solvers.direct_or_persistent_solver import (

DirectOrPersistentSolver,

)

from pyomo.common.enums import minimize, maximize

from pyomo.opt.results.results_ import SolverResults

from pyomo.opt.results.solution import Solution, SolutionStatus

from pyomo.opt.results.solver import TerminationCondition, SolverStatus

from pyomo.opt.base import SolverFactory

import time

import logging

import re

import time

from pyomo.common.collections import ComponentSet, ComponentMap, Bunch

from pyomo.common.dependencies import attempt_import

from pyomo.common.dependencies import numpy as np

from pyomo.core.base import Suffix, Var, Constraint, Objective

from pyomo.core.expr.numvalue import is_fixed

from pyomo.core.staleflag import StaleFlagManager

from pyomo.repn import generate_standard_repn

from pyomo.solvers.plugins.solvers.direct_solver import DirectSolver

from pyomo.solvers.plugins.solvers.direct_or_persistent_solver import (

DirectOrPersistentSolver,

)

from pyomo.common.enums import minimize, maximize

from pyomo.opt.results.results_ import SolverResults

from pyomo.opt.results.solution import Solution, SolutionStatus

from pyomo.opt.results.solver import TerminationCondition, SolverStatus

from pyomo.opt.base import SolverFactory

mrmundt · 2026-01-23T17:04:33Z

pyomo/solvers/plugins/solvers/cuopt_direct.py

+class CUOPTDirect(DirectSolver):
+    def __init__(self, **kwds):
+        kwds["type"] = "cuoptdirect"
+        super(CUOPTDirect, self).__init__(**kwds)


This is old-style Python. Suggest switching to new-style:

Suggested change

super(CUOPTDirect, self).__init__(**kwds)

super().__init__(**kwds)

pyomo/solvers/plugins/solvers/cuopt_direct.py

mrmundt · 2026-01-23T17:07:26Z

pyomo/solvers/plugins/solvers/cuopt_direct.py

+from pyomo.opt.base import SolverFactory
+import time
+
+logger = logging.getLogger(__name__)


This is never actually used anywhere. We love our loggers, but why bother declaring it if it's not used? Maybe you can change some of the error messages into log messages and/or add some helpful logging in between steps.

Utilized the logger to log intermediate metadata

- Remove unused imports - Use new-style super() - Add logging (debug for solve, warnings for MIP limitations)

Address requested changes

Iroy30 · 2026-01-29T03:22:04Z

@jdwit seems to have already handled the reviews. Thanks for jumping in! cc @mrmundt

add cuopt direct solver

90adbf4

blnicho requested a review from whart222 June 10, 2025 18:46

blnicho mentioned this pull request Jun 13, 2025

Support NVIDIA cuOpt solver #3626

Open

jsiirola added the AT: PRE-TEST INSPECTED label Jun 17, 2025

pyomo-autotest removed the AT: PRE-TEST INSPECTED label Jun 17, 2025

mrmundt reviewed Jun 17, 2025

View reviewed changes

Iroy30 added 2 commits July 2, 2025 05:29

add tests with lp amd milp capabilities

b4b1502

Merge remote-tracking branch 'origin/main'

6f64094

Iroy30 force-pushed the add_cuopt_direct_solver_plugin branch from 5dbf9dd to 6f64094 Compare July 2, 2025 12:33

mrmundt reviewed Jul 2, 2025

View reviewed changes

mrmundt added the AT: PRE-TEST INSPECTED label Jul 2, 2025

pyomo-autotest removed the AT: PRE-TEST INSPECTED label Jul 2, 2025

mrmundt added the AT: RETEST label Jul 9, 2025

pyomo-autotest removed the AT: RETEST label Jul 9, 2025

Iroy30 and others added 2 commits July 9, 2025 13:15

Merge branch 'main' into add_cuopt_direct_solver_plugin

d9e42ad

Update pyomo/solvers/plugins/solvers/cuopt_direct.py

e2e9159

Co-authored-by: Miranda Mundt <[email protected]>

mrmundt added the AT: PRE-TEST INSPECTED label Jul 9, 2025

pyomo-autotest removed the AT: PRE-TEST INSPECTED label Jul 9, 2025

mrmundt added the AT: RETEST label Jul 9, 2025

pyomo-autotest removed the AT: RETEST label Jul 9, 2025

Iroy30 added 3 commits July 9, 2025 16:47

Merge remote-tracking branch 'origin'

d4a5d07

formatting

61e19bf

pyomo/solvers/plugins/solvers/cuopt_direct.py

432d8c0

Don't perform redundant attempt_import('cuopt')

0e598ee

jsiirola added the AT: PRE-TEST INSPECTED label Nov 14, 2025

pyomo-autotest removed the AT: PRE-TEST INSPECTED label Nov 14, 2025

Iroy30 and others added 2 commits November 16, 2025 22:28

delete instance level version

134cba7

Update cuopt_direct.py

767d96a

jsiirola added the AT: PRE-TEST INSPECTED label Nov 17, 2025

pyomo-autotest removed the AT: PRE-TEST INSPECTED label Nov 17, 2025

pyomo-autotest added the AT: STALE label Dec 7, 2025

Iroy30 and others added 3 commits January 7, 2026 20:37

improve test coverage, add tests

6b84028

Merge branch 'main' into add_cuopt_direct_solver_plugin

982ef1e

Merge remote-tracking branch 'iroy30/add_cuopt_direct_solver_plugin'

eea17c0

pyomo-autotest removed the AT: STALE label Jan 8, 2026

Merge branch 'main' into add_cuopt_direct_solver_plugin

8fdf47c

jsiirola added the AT: PRE-TEST INSPECTED label Jan 9, 2026

pyomo-autotest removed the AT: PRE-TEST INSPECTED label Jan 9, 2026

Merge branch 'main' into add_cuopt_direct_solver_plugin

06f969b

blnicho added 2 commits January 20, 2026 14:08

Merge branch 'main' into add_cuopt_direct_solver_plugin

ce29917

Merge branch 'main' into add_cuopt_direct_solver_plugin

ed0d4bb

blnicho added the AT: PRE-TEST INSPECTED label Jan 22, 2026

pyomo-autotest removed the AT: PRE-TEST INSPECTED label Jan 22, 2026

mrmundt requested changes Jan 23, 2026

View reviewed changes

jdwit and others added 3 commits January 28, 2026 21:00

Address review feedback

ca0825a

- Remove unused imports - Use new-style super() - Add logging (debug for solve, warnings for MIP limitations)

Merge pull request #1 from jdwit/fix/cuopt-improvements

9d46db1

Address requested changes

Merge branch 'main' into add_cuopt_direct_solver_plugin

f893546

	super(CUOPTDirect, self).__init__(**kwds)
	super().__init__(**kwds)

add cuopt direct solver #3620

Are you sure you want to change the base?

add cuopt direct solver #3620

Conversation

Iroy30 commented Jun 4, 2025 • edited by blnicho Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Fixes #3626

Summary/Motivation:

Changes proposed in this PR:

Legal Acknowledgement

Uh oh!

codecov bot commented Jun 17, 2025 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Codecov Report

Uh oh!

mrmundt left a comment

Choose a reason for hiding this comment

Uh oh!

mrmundt commented Jul 1, 2025

Uh oh!

Iroy30 commented Jul 2, 2025

Uh oh!

Uh oh!

Uh oh!

mrmundt Jul 2, 2025

Choose a reason for hiding this comment

Uh oh!

whart222 commented Jul 8, 2025

Uh oh!

mrmundt commented Jul 8, 2025

Uh oh!

mrmundt commented Jul 8, 2025

Uh oh!

mrmundt commented Jul 8, 2025

Uh oh!

jsiirola commented Nov 14, 2025

Uh oh!

Iroy30 commented Nov 17, 2025 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

Iroy30 commented Nov 19, 2025

Uh oh!

jdwit commented Nov 24, 2025

Uh oh!

Iroy30 commented Jan 14, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

mrmundt Jan 23, 2026

Choose a reason for hiding this comment

Uh oh!

Iroy30 Jan 29, 2026

Choose a reason for hiding this comment

Uh oh!

mrmundt Jan 23, 2026

Choose a reason for hiding this comment

Uh oh!

Iroy30 Jan 29, 2026

Choose a reason for hiding this comment

Uh oh!

Uh oh!

mrmundt Jan 23, 2026

Choose a reason for hiding this comment

Uh oh!

Iroy30 Jan 29, 2026

Choose a reason for hiding this comment

Uh oh!

Iroy30 commented Jan 29, 2026

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

9 participants

Iroy30 commented Jun 4, 2025 •

edited by blnicho

Loading

codecov bot commented Jun 17, 2025 •

edited

Loading

Iroy30 commented Nov 17, 2025 •

edited

Loading

Iroy30 commented Jan 14, 2026 •

edited

Loading