ibragim-badertdinov/swe-rebench-07-2026
Coding
111 hard SWE-rebench leaderboard tasks from the July 2026 snapshot across Go, Java, Python, Rust, and TypeScript.
Run this task
CLI:
inspect eval inspect_harbor/swe_rebench_07_2026 --model openai/gpt-5Python:
from inspect_ai import eval
from inspect_harbor import swe_rebench_07_2026
eval(swe_rebench_07_2026(), model="openai/gpt-5")Dataset information
| Harbor registry | ibragim-badertdinov/swe-rebench-07-2026 |
| Inspect task | swe_rebench_07_2026 |
| Latest digest | sha256:e2e357045bf03e4900d2506c36562f6eaff7acd37f63780600967ea3aecdcd79 |
| Samples | 111 |
| Paper | arxiv |
| Source | https://huggingface.co/datasets/nebius/SWE-rebench-leaderboard |
See Task Parameters for the parameter set shared across all Harbor tasks.