-
Notifications
You must be signed in to change notification settings - Fork 63
Pull requests: OpenHands/benchmarks
Author
Label
Projects
Milestones
Reviews
Assignee
Sort
Pull requests list
fix: replace DelegateTool with TaskToolSet (removed in SDK v1.23.0)
#716
opened May 26, 2026 by
juanmichelini
Collaborator
Loading…
Harden workflow GitHub context handling
review-this
#714
opened May 24, 2026 by
enyst
Member
Loading…
DRAFT: swtbench: strip non-test files from model_patch in post-processing (option 2 of #708)
build-swt-bench
Build SWT-Bench images based on SDK version on this PR.
investigation
#711
opened May 13, 2026 by
juanmichelini
Collaborator
Loading…
DRAFT: swtbench: tighten default prompt to discourage non-test edits (option 1 of #708)
build-swt-bench
Build SWT-Bench images based on SDK version on this PR.
investigation
#710
opened May 13, 2026 by
juanmichelini
Collaborator
•
Draft
[DO NOT MERGE] ci(swtbench): revert to Blacksmith runner (fallback for #495)
#673
opened Apr 20, 2026 by
simonrosenberg
Collaborator
•
Draft
Fix GAIA build target docs: binary-minimal -> binary
#667
opened Apr 15, 2026 by
simonrosenberg
Collaborator
•
Draft
Remove redundant 'Apply workflow_dispatch overrides' step from build workflows
#660
opened Apr 13, 2026 by
simonrosenberg
Collaborator
•
Draft
Remove binary diffs from agent patches in SWE-bench evaluation
#656
opened Apr 9, 2026 by
juanmichelini
Collaborator
•
Draft
Filter SWE-Bench Multimodal image builds to curated subset
#644
opened Apr 6, 2026 by
juanmichelini
Collaborator
Loading…
fix: reset BuildKit cache between retries for base/assembly builds
#631
opened Apr 4, 2026 by
simonrosenberg
Collaborator
Loading…
3 tasks
Update Claude ACP package references
#629
opened Apr 3, 2026 by
simonrosenberg
Collaborator
Loading…
build(deps): bump the version-all group across 1 directory with 21 updates
dependencies
Pull requests that update a dependency file
python:uv
Pull requests that update python:uv code
#596
opened Mar 31, 2026 by
dependabot
Bot
Loading…
ProTip!
no:milestone will show everything without a milestone.