Dataset
SWEbench-Verified-eval150-M2.7-OPD149-visible-history-v2-2repeats-w64-20260921
by Charlie L CharlieLLL/SWEbench-Verified-eval150-M2.7-OPD149-visible-history-v2-2repeats-w64-20260921
User requested v2 only. Historical v1 score93/150 is retained as a non-matched reference: it preceded the Docker CPU environment fix.
Dataset Card
User requested v2 only. Historical v1 score93/150 is retained as a non-matched reference: it preceded the Docker CPU environment fix. New v2 runs use exactly the same OPD149 weights and model budgets with explicit history reasoning removal and CPU-quota-aware sandbox environments. Accuracy, M2.7 measured uncached input/cache reads/output and Docker-separated latency are primary outcomes. No new v1 full150 is required. Two repeats are screening evidence, not proof of non-inferiority.
Excerpt from the card by Charlie L.
Details
- Repository
- CharlieLLL/SWEbench-Verified-eval150-M2.7-OPD149-visible-history-v2-2repeats-w64-20260921
- Publisher
- Charlie L
- Task category
- Not stated by the source
- Tags
- Not stated by the source
- Size category
- Not stated by the source
- Languages
- Not stated by the source
- Revision
- 968ddecb12d1fb32eeb3cd1ec1ac1f1157014538
- Last updated
- 2026-09-21
Files
2 files, 3.1 KB in total.
Every file
| File | Type | Size | SHA-256 |
|---|---|---|---|
| README.md | Documentation | 567 B | — |
| .gitattributes | Repository | 2.5 KB | — |
License and Download
- License
- Not stated by the source
- Access
- No access gate
Released by Charlie L through its official repository on Hugging Face.