Cryptocurrency is a digital asset that continues to gain popularity due to its decentralized nature and potential for profit, but its high price volatility poses significant challenges for investors. This study aims to develop a price prediction software for Ethereum cryptocurrency using the Support Vector Regression (SVR) algorithm. Historical price data were collected, preprocessed, normalized using MinMaxScaler, and divided into training and testing datasets. The SVR model was optimized using the GridSearch method to obtain the best hyperparameters. Model performance was evaluated using MAE, RMSE, and MAPE, resulting in 199.61 (7.60%), 227.57 (8.66%), and 8.64%, respectively, indicating good predictive accuracy. The software was developed with the Flask framework and tested using Blackbox testing and stress testing via Locust, showing stable system performance with efficient response time. The developed software can serve as a decision-support tool for investors to predict Ethereum prices over various time ranges from 1 to 30 days or more