Background: The triglyceride glucose index (TyG-i), a biomarker of insulin resistance, has been associated with adverse vascular outcomes and risk stratification in several cardiovascular and cerebrovascular phenotypes. However, data on TyG-i as a prognostic marker in spontaneous intracerebral hemorrhage (ICH) remain limited. Objective: To explore the association between TyG-i and 90-day functional outcome in patients with ICH. Methods: A retrospective analysis of adult patients admitted for non-traumatic small vessel disease-related ICH at three Italian neurological institutions was conducted. TyG-i was calculated on admission as Ln[(fast triglycerides (mg/dL) × fast glucose (mg/dL))]/2. Functional outcome was measured with the modified Rankin Scale (mRS) at 90 days from the index event. TyG-i was analyzed as a continuous variable and categorized in quintiles (Q1 to Q5). Predictors of poor outcome (mRS 4–6) were investigated with multivariable logistic regression. Results: A total of 463 patients were included, of whom 197 (42.5%) had poor outcome at 90 days. TyG-i analyzed as a continuous variable was not associated with unfavorable prognosis. TyG-i analyzed as a categorical variable stratified by quintiles showed a non-linear U-shaped relationship with functional outcome; patients in Q4 had the lowest risk of poor outcome (Q1 reference, OR 0.44, 95% CI 0.22–0.87, p = 0.019). Discussion: We found a potential non-linear relationship between TyG-i and ICH outcome, with higher odds of good prognosis in patients with intermediate values. Conclusions: TyG-i may represent a promising, low-cost and widely available biomarker that might improve prognostication in clinical practice, but further studies are needed.