Your question is Cross-Validation for Model Selection. Take a moment with it on the right.
Talk me through your thinking if you like. When you're confident, submit your answer and I'll grade it like a real screen (7/10 or better passes).
You have several candidate models and one validation split is giving you inconsistent results. You want a cleaner way to compare them before picking a final model.
What is the role of cross-validation in model selection?