swebench_multilingual@1.0
Coding
SWE-bench Multilingual extends the original Python-focused SWE-bench benchmark to support multiple programming languages.
Run this task
CLI:
inspect eval inspect_harbor/swebench_multilingual_1_0 --model openai/gpt-5Python:
from inspect_ai import eval
from inspect_harbor import swebench_multilingual_1_0
eval(swebench_multilingual_1_0(), model="openai/gpt-5")Dataset information
| Harbor registry | swebench_multilingual@1.0 |
| Inspect task | swebench_multilingual_1_0 |
| Version | 1.0 |
| Samples | 300 |
| Paper | arxiv |
See Task Parameters for the parameter set shared across all Harbor tasks.