specific method · filed under evaluation
Harbor
Harbor is an evaluation harness identified as one of the tools used to run evaluations.
- sources
- 2
- models
- 2
- labs adopt it
- 2
- strongest
- used
How sources treat it
One count per evidence span, weakest treatment to strongest.
used 2
Documented in
Evidence
2 spans quoted from the sources, strongest treatment first.
using the Laude Institute's Harbor Framework with our agent harness
usedevaluation onlyin Laguna S 2.1 and Laguna XS 2.1Poolside
We used three main evaluation harnesses: Nemo Gym, Nemo Skills, and Harbor
usedevaluation onlyin Nemotron 3 UltraNVIDIA
Filed alongside
Other methods under evaluation :: evaluation harness.
DeepSeek Harness Minimal modeNeMo Evaluator SDKAtomic binary rubric evaluationAutomatic in-flight checkpoint evaluation schedulingBenchmark evaluation frameworkCross-scaffold evaluationInline training evaluatorsIsolated container per rolloutJSON-structured answer-format promptmini-swe-agent harnessNeMo Gym and NeMo Evaluator-based harnessNeMo SkillsPool