otel-bench@1.0
Coding
OpenTelemetry Benchmark - evaluates AI agents’ ability to instrument applications with OpenTelemetry tracing across multiple languages.
Run this task
CLI:
inspect eval inspect_harbor/otel_bench_1_0 --model openai/gpt-5Python:
from inspect_ai import eval
from inspect_harbor import otel_bench_1_0
eval(otel_bench_1_0(), model="openai/gpt-5")Dataset information
| Harbor registry | otel-bench@1.0 |
| Inspect task | otel_bench_1_0 |
| Version | 1.0 |
| Samples | 26 |
| Source | https://github.com/QuesmaOrg/otel-bench |
See Task Parameters for the parameter set shared across all Harbor tasks.