Skip to content

Update link from cuOpt 26.06 to 26.08 - #10

Merged
0x17 merged 5 commits into
mainfrom
cuopt-v26.08
Aug 7, 2026
Merged

Update link from cuOpt 26.06 to 26.08#10
0x17 merged 5 commits into
mainfrom
cuopt-v26.08

Conversation

@0x17

@0x17 0x17 commented Aug 7, 2026

Copy link
Copy Markdown
Member
  • Updated Dependencies: Bumped cuopt-cu12 and cuopt-cu13 dependencies to 26.8.* across x86_64 and ARM64 workflows.
  • Parameter Safety: Filtered out the removed solve-by-pdlp parameter in gmscuopt.c with a warning log to prevent runtime option errors.
  • Quadratic Maximization Alignment: Restricted dual variable sign inversion (*= -1.0) strictly to LP models, leveraging cuOpt 26.08's native QP maximization dual handling.
  • CI improvements: Switched the test runner to uv and upgraded CI Python from 3.12 to 3.13.

@0x17 0x17 self-assigned this Aug 7, 2026
@0x17 0x17 added the cuopt-release Update the link for a new NVIDIA cuopt release label Aug 7, 2026
@0x17
0x17 marked this pull request as ready for review August 7, 2026 08:02
@0x17
0x17 marked this pull request as draft August 7, 2026 08:03
@0x17
0x17 marked this pull request as ready for review August 7, 2026 08:30
@0x17
0x17 merged commit dfe6167 into main Aug 7, 2026
2 checks passed
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

cuopt-release Update the link for a new NVIDIA cuopt release

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant