Free online curve fitting in your browser: pick a function — Gaussian, Lorentzian, exponential decay, logistic and more — and fit it to your measurement data with least squares nonlinear regression. Upload your CSV, set start values and bounds, and get fitted parameters with uncertainties and R² in seconds — nothing to install.
Straight lines and gentle curvature for trends and calibrations.
Bell-shaped peaks for spectra, chromatography and resonances.
Periodic signals and damped oscillations.
Exponential rise, relaxation and saturation processes.
S-shaped transitions between two levels.
Scaling laws and logarithmic growth (x > 0).
Curve fitting means finding the function that best describes your measurement data. Sci-eval does it with least squares nonlinear regression: it adjusts each parameter of the function you choose until the squared distance between the fitted curve and your data points is as small as possible. Everything runs online — there is nothing to install.
Start from a model above — a Gaussian or Lorentzian peak for spectra, an exponential decay for kinetics, a logistic or Hill curve for dose–response, a power law for scaling behaviour — or define your own function in SymPy syntax. Upload a CSV, map the x and y columns, and set start values and bounds to guide the optimiser. Sci-eval returns the fitted parameters with their uncertainties and an R² goodness-of-fit, plus a plot of the fitted curve over your data that you can export for reports and publications.
Online curve fitting is the process of finding the function whose parameters best reproduce your measured data, right in a web browser. Sci-eval uses least squares nonlinear regression: it adjusts the parameters until the squared distance between the fitted curve and your data points is as small as possible.
Choose a function from the library above, upload your data as a CSV, and map the x and y columns. Set start values and bounds to guide the optimiser, then run the fit — you get the fitted parameters with uncertainties and an R² goodness-of-fit within seconds.
The standard library covers Gaussian and Lorentzian peaks, exponential growth, decay and saturation, logistic, Gompertz and Hill dose–response curves, power laws, and logarithmic and polynomial trends. You can also define your own model in SymPy syntax for a fully custom fit.
Yes. Curve fitting on Sci-eval is free — register for a free account and your data and fit results are saved to your workspace so you can revisit and export them.
Every fit returns the optimised parameter values with their standard-error uncertainties, an R² goodness-of-fit, and a plot of the fitted curve over your data that you can export for reports and publications.