6 Commits

Author SHA1 Message Date
sora
a37a1165b2 Preload Pier uv and mini-swe-agent wheels so DeepSWE sandbox builds skip GitHub.
Co-authored-by: Cursor <cursoragent@cursor.com>
2026-09-02 09:44:17 +00:00
sora
ee54cd6d5d Rename SciCode preload script away from the K3 sandbox name.
Co-authored-by: Cursor <cursoragent@cursor.com>
2026-09-02 08:41:09 +00:00
sora
70fe0c1d95 Build SciCode sandbox image from local wheels to avoid PyPI timeouts.
Co-authored-by: Cursor <cursoragent@cursor.com>
2026-09-02 08:08:46 +00:00
sora
d85e9986c8 Add DeepSWE task image preload script.
Co-authored-by: Cursor <cursoragent@cursor.com>
2026-09-01 03:36:30 +00:00
sora
0f7c460f1b Add SciCode sandbox preload and extra bench configs for dpv4.
Keep existing yaml entries unchanged; enable SciCode sandbox in run.py and host-side image build without GDPval.

Co-authored-by: Cursor <cursoragent@cursor.com>
2026-08-31 07:24:51 +00:00
sora
ba2b5c1772 fix: terminal_bench_v2_1 result collection, apt mirror, and restore run.py env var 2026-08-05 03:56:44 +00:00