NVIDIA cuOpt - Browse /v26.08.00 at SourceForge.net

The interactive file manager requires Javascript. Please enable it or use sftp or scp.
You may still browse the files here.

Name	Modified	Size	InfoDownloads / Week
Parent folder
README.md	2026-08-06	22.3 kB	0
v26.08.00 source code.tar.gz	2026-08-06	45.6 MB	0
v26.08.00 source code.zip	2026-08-06	46.5 MB	0
Totals: 3 Items		92.1 MB	0

What's Changed

🚨 Breaking Changes

Add short cu12/cu13 Docker tag aliases and update docs by @ramakrishnap-nv in https://github.com/NVIDIA/cuopt/pull/1370
skills: consolidate three numerical-optimization API skills into one by @ramakrishnap-nv in https://github.com/NVIDIA/cuopt/pull/1489
wheels: build CUDA 13 wheels with latest CTK (13.3.0) by @jameslamb in https://github.com/NVIDIA/cuopt/pull/1527
Routing DataModel: store-then-build (defer device construction to solve) by @ramakrishnap-nv in https://github.com/NVIDIA/cuopt/pull/1556
Remove deprecated solve-by-pdlp attribute in Python API by @nguidotti in https://github.com/NVIDIA/cuopt/pull/1664

🐛 Bug Fixes

Fix mixed-dtype cudf.concat in test_initial_solutions helper by @ramakrishnap-nv in https://github.com/NVIDIA/cuopt/pull/1295
check for valid size in cuOptGetErrorString by @mlubin in https://github.com/NVIDIA/cuopt/pull/1414
Fix PDLP capture cublas error and hang on infeasible solutions by @akifcorduk in https://github.com/NVIDIA/cuopt/pull/1416
Add a guard for huge bounds in bounds propagation by @hlinsen in https://github.com/NVIDIA/cuopt/pull/1390
grpc fix a race condition with log streaming by @tmckayus in https://github.com/NVIDIA/cuopt/pull/1426
bugfix (routing): fix min_vehicles bug by @np96 in https://github.com/NVIDIA/cuopt/pull/1405
fix(routing): lower opt level for sliding_window.cu to avoid nvcc 13.0.3 ICE by @ramakrishnap-nv in https://github.com/NVIDIA/cuopt/pull/1452
Assign gRPC workers to distinct GPUs via cudaSetDevice by @tmckayus in https://github.com/NVIDIA/cuopt/pull/1449
ci: fix unbound CONDA_PREFIX in test_doc_examples.sh by @ramakrishnap-nv in https://github.com/NVIDIA/cuopt/pull/1462
Fix the bug detecting nonconvex quadratic constraints in the general … by @yuwenchen95 in https://github.com/NVIDIA/cuopt/pull/1438
fix(ci): fix typo in notebook skip list for transport_cuopt by @ramakrishnap-nv in https://github.com/NVIDIA/cuopt/pull/1496
fix: validate row type byte before enum cast in MPS parser by @ramakrishnap-nv in https://github.com/NVIDIA/cuopt/pull/1491
Fix root cut CPUFJ solutions being dropped by the GPU heuristics by @aliceb-nv in https://github.com/NVIDIA/cuopt/pull/1497
fix(socp): canonicalize QC Q COO for rotated SOC detection by @yuwenchen95 in https://github.com/NVIDIA/cuopt/pull/1439
fix: incorrect GF2 presolve constraint addressing by @aliceb-nv in https://github.com/NVIDIA/cuopt/pull/1517
Remove wrong exception of unsupported QCQP by @rg20 in https://github.com/NVIDIA/cuopt/pull/1493
fix(barrier): destroy cuDSS descriptors before freeing their backing buffers by @ramakrishnap-nv in https://github.com/NVIDIA/cuopt/pull/1555
fix(clang-format): preserve SPDX comments by @bdice in https://github.com/NVIDIA/cuopt/pull/1560
fix(ci): exclude cancelled jobs from pr-test-summary failed count by @ramakrishnap-nv in https://github.com/NVIDIA/cuopt/pull/1558
Fix cuts not obeying timer by @akifcorduk in https://github.com/NVIDIA/cuopt/pull/1563
[BUG] Fix variable fixing for initial solutions by @nguidotti in https://github.com/NVIDIA/cuopt/pull/1580
Fix clique size and numerical issue bugs by @akifcorduk in https://github.com/NVIDIA/cuopt/pull/1589
fix(build): join EXTRA_CMAKE_ARGS with semicolons for SKBUILD_CMAKE_ARGS by @ramakrishnap-nv in https://github.com/NVIDIA/cuopt/pull/1588
Fix libomp ABI bug by @aliceb-nv in https://github.com/NVIDIA/cuopt/pull/1593
make objective_scaling_factor optional for gRPC by @tmckayus in https://github.com/NVIDIA/cuopt/pull/1600
Fix SOC index collision and simplify QC-to-SOC conversion by routing linear QCs through general LDLT path by @yuwenchen95 in https://github.com/NVIDIA/cuopt/pull/1564
fix(grpc): cancel active jobs on delete by @tmckayus in https://github.com/NVIDIA/cuopt/pull/1604
Fix Ruiz equilibration skip heuristic to also check column imbalance by @mlubin in https://github.com/NVIDIA/cuopt/pull/1601
fix(routing): preserve row-major layout when resizing capacity routes by @mlubin in https://github.com/NVIDIA/cuopt/pull/1621
fix(docker): pre-install test deps so dnf removal doesn't break test_image.sh by @ramakrishnap-nv in https://github.com/NVIDIA/cuopt/pull/1627
Allow zero-valued LP Python model updates by @fallintoplace in https://github.com/NVIDIA/cuopt/pull/1421
fix(grpc): terminate workers on server shutdown by @tmckayus in https://github.com/NVIDIA/cuopt/pull/1603
test(barrier): cap qplib_8515_ruiz_forced_off at the iteration limit by @mlubin in https://github.com/NVIDIA/cuopt/pull/1628
Fix routing YAML best results export by @fallintoplace in https://github.com/NVIDIA/cuopt/pull/1422
Support maximization in quadratic programs by @chris-maes in https://github.com/NVIDIA/cuopt/pull/1641
modify ldconfig in ubi image so nvidia libraries are on path by @tmckayus in https://github.com/NVIDIA/cuopt/pull/1644
Temporarily exclude GB300 from testing by @KyleFromNVIDIA in https://github.com/NVIDIA/cuopt/pull/1648
Revert "Temporarily exclude GB300 from testing (#1648)" by @KyleFromNVIDIA in https://github.com/NVIDIA/cuopt/pull/1651
remove extra or unnecessary parameters in Python gRPC api by @tmckayus in https://github.com/NVIDIA/cuopt/pull/1654
Pin PaPILO to flush empty-row fix from akifcorduk/papilo. by @akifcorduk in https://github.com/NVIDIA/cuopt/pull/1645
Fixed 26.08 release notes by @nguidotti in https://github.com/NVIDIA/cuopt/pull/1665

📖 Documentation

Update cuOpt MIP positioning by @Sylendran95 in https://github.com/NVIDIA/cuopt/pull/1387
Fix FAQ command typos in health-check examples by @jackthepunished in https://github.com/NVIDIA/cuopt/pull/1389
docs(grpc): title-case headings, replace ASCII diagrams with PNGs, archive older version URLs by @ramakrishnap-nv in https://github.com/NVIDIA/cuopt/pull/1420
added kaminpar license to THIRD_PARTY_LICENSES by @Bubullzz in https://github.com/NVIDIA/cuopt/pull/1463
docs: fix NGC container catalog URL by @ramakrishnap-nv in https://github.com/NVIDIA/cuopt/pull/1470
Align cuOpt documentation branding and Doxygen configuration by @bdice in https://github.com/NVIDIA/cuopt/pull/1503
normalize MILP to MIP in documentation by @tmckayus in https://github.com/NVIDIA/cuopt/pull/1655
docs: add Python async gRPC client guide under gRPC remote execution by @tmckayus in https://github.com/NVIDIA/cuopt/pull/1653
docs: release notes for 26.08 by @ramakrishnap-nv in https://github.com/NVIDIA/cuopt/pull/1638

🚀 New Features

add a Python interface to the C++ grpc asyc client by @tmckayus in https://github.com/NVIDIA/cuopt/pull/1430
Implement multi-GPU PDLP by @Bubullzz in https://github.com/NVIDIA/cuopt/pull/1391
Add support for sparsity exploitation for SOCs of large dimensionality by @yuwenchen95 in https://github.com/NVIDIA/cuopt/pull/1468
Make the post solve info logging optional by @rg20 in https://github.com/NVIDIA/cuopt/pull/1639

🛠️ Improvements

Fix version failure in check style for skills by @ramakrishnap-nv in https://github.com/NVIDIA/cuopt/pull/1271
Main merge release/26.06.01 by @ramakrishnap-nv in https://github.com/NVIDIA/cuopt/pull/1273
Main merge release/26.06 03 by @ramakrishnap-nv in https://github.com/NVIDIA/cuopt/pull/1296
Fix skill-validation failures across all skills by @ramakrishnap-nv in https://github.com/NVIDIA/cuopt/pull/1287
skills: rename resources/ to references/ by @ramakrishnap-nv in https://github.com/NVIDIA/cuopt/pull/1301
skills: add evals/evals.json smoke suite (group A) by @ramakrishnap-nv in https://github.com/NVIDIA/cuopt/pull/1308
skills: add evals/evals.json smoke suite by @ramakrishnap-nv in https://github.com/NVIDIA/cuopt/pull/1302
Main merge release/26.06 05 by @ramakrishnap-nv in https://github.com/NVIDIA/cuopt/pull/1318
skills/cuopt-install:Add eval by @ramakrishnap-nv in https://github.com/NVIDIA/cuopt/pull/1324
skills/cuopt-numerical-optimization-api-c: Add eval by @ramakrishnap-nv in https://github.com/NVIDIA/cuopt/pull/1327
Rename routing-formulation to cuopt-routing-formulation by @ramakrishnap-nv in https://github.com/NVIDIA/cuopt/pull/1329
Rename numerical-optimization-formulation to cuopt-numerical-optimization-formulation by @ramakrishnap-nv in https://github.com/NVIDIA/cuopt/pull/1328
skills/cuopt-server-common: Add eval by @ramakrishnap-nv in https://github.com/NVIDIA/cuopt/pull/1325
Rename skill-evolution to cuopt-skill-evolution by @ramakrishnap-nv in https://github.com/NVIDIA/cuopt/pull/1330
skills/cuopt-server-api-python: update with benchmark details by @ramakrishnap-nv in https://github.com/NVIDIA/cuopt/pull/1334
skills/cuopt-numerical-optimization-api-python: update with benchmark details by @ramakrishnap-nv in https://github.com/NVIDIA/cuopt/pull/1332
skills/cuopt-routing-api-python: update with benchmark details by @ramakrishnap-nv in https://github.com/NVIDIA/cuopt/pull/1333
skills/cuopt-user-rules: update with benchmark details by @ramakrishnap-nv in https://github.com/NVIDIA/cuopt/pull/1335
skills/cuopt-numerical-optimization-api-cli: update with benchmark details by @ramakrishnap-nv in https://github.com/NVIDIA/cuopt/pull/1331
skills/cuopt-skill-evolution: add eval by @ramakrishnap-nv in https://github.com/NVIDIA/cuopt/pull/1337
skills: prep for nvidia/skills catalog onboarding by @ramakrishnap-nv in https://github.com/NVIDIA/cuopt/pull/1310
Main merge release/26.06 11 by @ramakrishnap-nv in https://github.com/NVIDIA/cuopt/pull/1360
fix(ci): fix configuration for breaking change notification workflow by @jameslamb in https://github.com/NVIDIA/cuopt/pull/1356
Main merge release/26.06 12 by @ramakrishnap-nv in https://github.com/NVIDIA/cuopt/pull/1367
Main merge release/26.06 14 by @ramakrishnap-nv in https://github.com/NVIDIA/cuopt/pull/1374
Add cuopt-multi-objective-exploration skill (Pareto frontier) by @cafzal in https://github.com/NVIDIA/cuopt/pull/1355
Remove CUDA whole compilation programming model violations by @robertmaynard in https://github.com/NVIDIA/cuopt/pull/1347
ci: nudge multi-objective skill to trigger eval by @ramakrishnap-nv in https://github.com/NVIDIA/cuopt/pull/1392
Note dual/sensitivity support per problem type in formulation skill by @cafzal in https://github.com/NVIDIA/cuopt/pull/1393
Expose diving hyper parameters by @nguidotti in https://github.com/NVIDIA/cuopt/pull/1364
Fix adj_list check bug and improve clique cuts by @akifcorduk in https://github.com/NVIDIA/cuopt/pull/1386
evals: add benchmark evals for cuopt-developer skill by @ramakrishnap-nv in https://github.com/NVIDIA/cuopt/pull/1399
Use LP format reader throughout unit tests by @mlubin in https://github.com/NVIDIA/cuopt/pull/1410
Correct dual-value scope to LP/QP in numerical-optimization API skills by @cafzal in https://github.com/NVIDIA/cuopt/pull/1408
evals: add 3 C API evals for cuopt-numerical-optimization-api-c skill by @ramakrishnap-nv in https://github.com/NVIDIA/cuopt/pull/1417
refactor: switch to rapids-artifact-name for consistent artifact naming by @gforsyth in https://github.com/NVIDIA/cuopt/pull/1411
evals: add 3 Python API evals for cuopt-numerical-optimization-api-python skill by @ramakrishnap-nv in https://github.com/NVIDIA/cuopt/pull/1418
ci: update nightly run branch from release/26.06 to release/26.08 by @ramakrishnap-nv in https://github.com/NVIDIA/cuopt/pull/1424
Vector length diving + Farkas diving by @nguidotti in https://github.com/NVIDIA/cuopt/pull/1401
Remove pinned host memory from barrier solver by @rg20 in https://github.com/NVIDIA/cuopt/pull/1321
Read epsilon-constraint duals as frontier slope / exchange rate in multi-objective skill by @cafzal in https://github.com/NVIDIA/cuopt/pull/1406
Build and test with CUDA 13.3.0 by @bdice in https://github.com/NVIDIA/cuopt/pull/1409
skills(multi-objective): add blank line after frontmatter to enable skill signing by @ramakrishnap-nv in https://github.com/NVIDIA/cuopt/pull/1433
evals: add 3 formulation evals for cuopt-numerical-optimization-formulation skill by @ramakrishnap-nv in https://github.com/NVIDIA/cuopt/pull/1398
skills: allow user-space env setup; refuse only sudo/system-level ops by @ramakrishnap-nv in https://github.com/NVIDIA/cuopt/pull/1448
Correct quadratic-constraint claims in formulation skill by @cafzal in https://github.com/NVIDIA/cuopt/pull/1407
Use main shared-workflows branch by @bdice in https://github.com/NVIDIA/cuopt/pull/1455
ci: disable pulp test_infeasible_2 pending fix in [#1115] by @ramakrishnap-nv in https://github.com/NVIDIA/cuopt/pull/1454
ci: trigger numerical optimization skill evals by @ramakrishnap-nv in https://github.com/NVIDIA/cuopt/pull/1453
Restore resource_cast memory limit check by @bdice in https://github.com/NVIDIA/cuopt/pull/1231
ci: reduce wheel-tests-cuopt-server matrix to 1 job per CUDA major by @ramakrishnap-nv in https://github.com/NVIDIA/cuopt/pull/1457
Add latent-objective recognition eval to multi-objective skill by @cafzal in https://github.com/NVIDIA/cuopt/pull/1442
Update CuPy and NumPy dependency specs by @bdice in https://github.com/NVIDIA/cuopt/pull/1456
skills(multi-objective): trigger re-eval after adding eval-005 by @ramakrishnap-nv in https://github.com/NVIDIA/cuopt/pull/1460
ci: remove dataset downloads from Python test scripts by @ramakrishnap-nv in https://github.com/NVIDIA/cuopt/pull/1458
chore(ci): exclude NVSkills-generated files from trailing-whitespace hook by @ramakrishnap-nv in https://github.com/NVIDIA/cuopt/pull/1465
Refactor: mathematical_optimization namespace by @mlubin in https://github.com/NVIDIA/cuopt/pull/1446
build(deps): bump msgpack from 1.1.2 to 1.2.1 by @ramakrishnap-nv in https://github.com/NVIDIA/cuopt/pull/1464
Post sticky PR comment summarizing CI test job failures by @ramakrishnap-nv in https://github.com/NVIDIA/cuopt/pull/1412
skills(cuopt-user-rules): update for mathematical_optimization namespace refactor by @ramakrishnap-nv in https://github.com/NVIDIA/cuopt/pull/1480
skills(cuopt-numerical-optimization-formulation): update for mathematical_optimization namespace refactor by @ramakrishnap-nv in https://github.com/NVIDIA/cuopt/pull/1479
skills(cuopt-numerical-optimization-api-c): update for mathematical_optimization namespace refactor by @ramakrishnap-nv in https://github.com/NVIDIA/cuopt/pull/1478
skills(cuopt-developer): update for mathematical_optimization namespace refactor by @ramakrishnap-nv in https://github.com/NVIDIA/cuopt/pull/1475
skills(cuopt-multi-objective-exploration): update for mathematical_optimization namespace refactor by @ramakrishnap-nv in https://github.com/NVIDIA/cuopt/pull/1477
skills(cuopt-install): update for mathematical_optimization namespace refactor by @ramakrishnap-nv in https://github.com/NVIDIA/cuopt/pull/1476
MIP log cleanup by @nguidotti in https://github.com/NVIDIA/cuopt/pull/1402
Zero Half(Odd-Cycle) cuts by @akifcorduk in https://github.com/NVIDIA/cuopt/pull/1428
ci: remove lean CI matrix from pr.yaml by @ramakrishnap-nv in https://github.com/NVIDIA/cuopt/pull/1474
skills: remove redundant concept-only skills by @ramakrishnap-nv in https://github.com/NVIDIA/cuopt/pull/1487
Guard CuPy CTK extra for wheel metadata by @bdice in https://github.com/NVIDIA/cuopt/pull/1469
refactor: promote shared primitives to mathematical_optimization by @mlubin in https://github.com/NVIDIA/cuopt/pull/1481
test: clean up gRPC server process groups by @mlubin in https://github.com/NVIDIA/cuopt/pull/1492
Improve cuopt-numerical-optimization-api-cli by @mlubin in https://github.com/NVIDIA/cuopt/pull/1495
ci(test): enable pytest-xdist for cuopt Python tests by @ramakrishnap-nv in https://github.com/NVIDIA/cuopt/pull/1486
Papilo primal/dual crush by @aliceb-nv in https://github.com/NVIDIA/cuopt/pull/1104
Disable coverage in cuOpt.jl tests by @mlubin in https://github.com/NVIDIA/cuopt/pull/1504
Update conventions for container indexing and casting by @mlubin in https://github.com/NVIDIA/cuopt/pull/1505
feat(docker): add UBI10 container image for FIPS compliance by @ramakrishnap-nv in https://github.com/NVIDIA/cuopt/pull/1499
fix(docker): use Docker Hub for UBI10 CUDA base images by @ramakrishnap-nv in https://github.com/NVIDIA/cuopt/pull/1511
fix(ci): restrict UBI10 build, test, and manifest to CUDA 13.x only by @ramakrishnap-nv in https://github.com/NVIDIA/cuopt/pull/1515
fix(grpc): drain all remaining log lines at job completion in StreamLogs by @ramakrishnap-nv in https://github.com/NVIDIA/cuopt/pull/1508
ci: add multi-GPU (NCCL) cpp test job on 2-GPU runner by @ramakrishnap-nv in https://github.com/NVIDIA/cuopt/pull/1519
style(cpp): drop gratuitous static_cast<size_t> on container subscripts by @mlubin in https://github.com/NVIDIA/cuopt/pull/1506
Constrain server data file paths by @fallintoplace in https://github.com/NVIDIA/cuopt/pull/1423
ci(test): fix grpc server GPU-memory leak and run cuopt tests in parallel by @ramakrishnap-nv in https://github.com/NVIDIA/cuopt/pull/1512
Update RAPIDS.cmake to log source of rapids-cmake by @arhag23 in https://github.com/NVIDIA/cuopt/pull/1521
remove 'cuda-python' dependency, test against oldest numba and numpy by @jameslamb in https://github.com/NVIDIA/cuopt/pull/1520
Empty commit to trigger new builds due to RAFT 3052 ABI breakage by @divyegala in https://github.com/NVIDIA/cuopt/pull/1542
make selection of CPU or GPU memory for problem automatic in C api by @tmckayus in https://github.com/NVIDIA/cuopt/pull/1522
Accept numpy and pandas inputs in routing Python API by @ramakrishnap-nv in https://github.com/NVIDIA/cuopt/pull/1546
Add a log-once macro by @aliceb-nv in https://github.com/NVIDIA/cuopt/pull/1545
Removed b_bar test for Gomory Cuts by @nguidotti in https://github.com/NVIDIA/cuopt/pull/1544
Python grpc async api allow arguments for TLS by @tmckayus in https://github.com/NVIDIA/cuopt/pull/1525
cuopt-skill-evolution: add low-maintenance/fail-loud design rule by @ramakrishnap-nv in https://github.com/NVIDIA/cuopt/pull/1561
Clamp crushed solutions to dual-reduced bounds by @aliceb-nv in https://github.com/NVIDIA/cuopt/pull/1568
Increase cuSparse version requierment for SpMVOp from 12.7 to 12.8 by @Bubullzz in https://github.com/NVIDIA/cuopt/pull/1576
Conflict Graph Improvements: Use non-binary rows for and add probing implications to CG by @akifcorduk in https://github.com/NVIDIA/cuopt/pull/1569
ci: enable codecov for Python test coverage by @ramakrishnap-nv in https://github.com/NVIDIA/cuopt/pull/1575
Recursive RINS by @nguidotti in https://github.com/NVIDIA/cuopt/pull/1482
Fast MPS parser for free-format MPS files by @aliceb-nv in https://github.com/NVIDIA/cuopt/pull/1429
Routing Phase 3: export the recorded problem to a host representation by @ramakrishnap-nv in https://github.com/NVIDIA/cuopt/pull/1583
change semantics of problem construction in cuOptReadProblem by @tmckayus in https://github.com/NVIDIA/cuopt/pull/1559
Make cuopt.routing importable without a GPU by @ramakrishnap-nv in https://github.com/NVIDIA/cuopt/pull/1591
Add front-completion workflow to cuopt-multi-objective-exploration skill by @cafzal in https://github.com/NVIDIA/cuopt/pull/1550
fix(deps): add cufile to test dependencies for cuda12 by @gforsyth in https://github.com/NVIDIA/cuopt/pull/1553
fix(build): pin RAPIDS_BRANCH to branch-26.08 by @ramakrishnap-nv in https://github.com/NVIDIA/cuopt/pull/1599
tests: retire correctness/regression tests covered by C++ gtests by @ramakrishnap-nv in https://github.com/NVIDIA/cuopt/pull/1461
fix(docker): patch UBI10 High CVEs — dnf update + remove vim by @ramakrishnap-nv in https://github.com/NVIDIA/cuopt/pull/1602
extend getters on C API by @tmckayus in https://github.com/NVIDIA/cuopt/pull/1549
fix(docker): patch High CVEs in UBI10 and Ubuntu container images by @ramakrishnap-nv in https://github.com/NVIDIA/cuopt/pull/1611
developer skills: size PARALLEL_LEVEL from cgroup limits in containers by @mlubin in https://github.com/NVIDIA/cuopt/pull/1614
Print logs when exception is caught by @akifcorduk in https://github.com/NVIDIA/cuopt/pull/1590
fix(quality): resolve SonarQube S3490 and S1186 critical findings by @ramakrishnap-nv in https://github.com/NVIDIA/cuopt/pull/1609
docs(skills): add GitHub etiquette rules and evals to cuopt-developer by @ramakrishnap-nv in https://github.com/NVIDIA/cuopt/pull/1619
refactor(tests): link internal tests against cuopt_static by @ramakrishnap-nv in https://github.com/NVIDIA/cuopt/pull/1581
enforce 'yamllint' checks by @jameslamb in https://github.com/NVIDIA/cuopt/pull/1613
test routing null matrix validation by @aycsi in https://github.com/NVIDIA/cuopt/pull/1220
fix(security): resolve SonarQube critical/high/blocker security findings by @ramakrishnap-nv in https://github.com/NVIDIA/cuopt/pull/1608
fix(docker): don't override python3 alternative in UBI10 image by @ramakrishnap-nv in https://github.com/NVIDIA/cuopt/pull/1634
test(mip): Update incumbent callback API test to use a bit more complex problem by @ramakrishnap-nv in https://github.com/NVIDIA/cuopt/pull/1649
Include Q matrix in Ruiz scale computations by @rg20 in https://github.com/NVIDIA/cuopt/pull/1643
skills(cuopt-multi-objective-exploration): trigger CI eval by @ramakrishnap-nv in https://github.com/NVIDIA/cuopt/pull/1642
Reduce concurrent LP host memory footprint by @hlinsen in https://github.com/NVIDIA/cuopt/pull/1640