specific method · filed under evaluation
DeepSeek Harness Minimal mode
This is the Minimal mode of DeepSeek Harness used as the agent evaluation framework.
Also called DeepSeek Harness Minimal evaluation mode, Minimal mode of DeepSeek Harness.
- sources
- 2
- models
- 2
- lab adopt it
- 1
- 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.
DeepSeek models are evaluated with the minimal mode of DeepSeek Harness as the agent framework
usedevaluation onlyin DeepSeek-V4-Flash-Vision-ExpDeepSeek
the model is evaluated with the Minimal mode of DeepSeek Harness and a 1M-token context window.
usedevaluation onlyin DeepSeek-V4.1-FlashDeepSeek
Filed alongside
Other methods under evaluation :: evaluation harness.
HarborNeMo 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