This paper studies stochastic volatility models under the price of EUR/USD exchange rate options traded in Lithuania. Three models were considered: Hull-White, Heston and logarithmical Ornstein-Uhlenbeck stochastic volatility model.The model performance is assessed by two criteria. First, in the sample fit of each model by comparing the implied volatility pattern generated either from the market price or from stochastic volatility models prices.Second, out of sample forecast performance is tested by comparing the one-day ahead forecasting accuracy of the Black-Scholes&German-Kohlhagen model with these of the stochastic volatility models.