The paper presents the development of a nonlinear least squares program called MULTI for microcomputers. written in BASIC, MULTI can fit up to five pharmacokinetic equations simultaneously to observed time courses. Four algorithms—Gauss-Newton, damping Gauss-Newton, modified Marquardt, and simplex—are available for curve fitting. The program is demonstrated using time courses of ampicillin and oxacillin in humans. MULTI is designed to be user-friendly, with detailed prompts for input data and provides output including estimated parameters, standard deviations, minimum SS, and AIC values. The simplex method, known for its robustness, is particularly useful when initial parameter values are far from the final converged values. However, it may stop calculation prematurely if SS reaches an extreme value. The Gauss-Newton methods are more efficient if initial values are close to the final values. The cooperative use of these algorithms is recommended for accurate results. The paper also discusses the application of Akaike's information criterion (AIC) for model selection and provides examples of MULTI's execution, including the fitting of ampicillin and oxacillin time courses.The paper presents the development of a nonlinear least squares program called MULTI for microcomputers. written in BASIC, MULTI can fit up to five pharmacokinetic equations simultaneously to observed time courses. Four algorithms—Gauss-Newton, damping Gauss-Newton, modified Marquardt, and simplex—are available for curve fitting. The program is demonstrated using time courses of ampicillin and oxacillin in humans. MULTI is designed to be user-friendly, with detailed prompts for input data and provides output including estimated parameters, standard deviations, minimum SS, and AIC values. The simplex method, known for its robustness, is particularly useful when initial parameter values are far from the final converged values. However, it may stop calculation prematurely if SS reaches an extreme value. The Gauss-Newton methods are more efficient if initial values are close to the final values. The cooperative use of these algorithms is recommended for accurate results. The paper also discusses the application of Akaike's information criterion (AIC) for model selection and provides examples of MULTI's execution, including the fitting of ampicillin and oxacillin time courses.