Make publication-quality graphs with real gnuplot 6 running inside your browser. Nothing to install, and your data never leaves your computer. Use Easy mode without writing a single command: choose a chart, paste your data from Excel, type your axis labels and download. If you already know gnuplot, switch to Gnuplot code and write scripts with a live preview.
Loading Gnuplot Online…
How to make a graph in 4 steps (no code)
- Choose a chart: line chart, scatter with a curve fit, bar chart, box plot, histogram, heat map or two y-axes.
- Add your data: copy the cells in Excel or Google Sheets and paste them into the first cell of the table, or upload a CSV file. Columns named SD, SE or SEM next to a data column become its error bars automatically.
- Pick the columns and labels: tick the series to plot, type the axis labels (use ^{2} for superscript and _{600} for subscript), and choose the legend position.
- Set the size and download: pick your journal width (Nature 89 or 183 mm, Cell 85, 114 or 174 mm, Science, PNAS, Elsevier, thesis or slide) and download PNG, PDF, SVG, EPS or a LaTeX file.
What you can do
- Curve fitting with standard errors: linear, quadratic, exponential, decay with half-life, power law, logistic growth, Michaelis–Menten (Vmax, Km) and dose–response 4PL (EC50, Hill slope). Each fit shows the parameters ± standard error, R² and degrees of freedom.
- Journal style in one click: open axes, outward ticks, thin lines and colour-blind-safe Okabe–Ito colours, or Nature-style, Lancet-style and greyscale palettes.
- 16 ready-made gnuplot scripts: growth curves, calibration lines, dose–response, enzyme kinetics, Arrhenius plots, bar charts with error bars, box plots, histograms, multi-panel figures (a–d), heat maps, 3D surfaces, Kaplan–Meier survival curves and PCA score plots.
- LaTeX output: pict2e (a single .tex file, text set in your document’s font) and epslatex (.tex + .eps). Download the script or a project zip with your data, a Makefile and a .gitignore, so you can keep only text files under version control.
Frequently asked questions
Is this real gnuplot?
Yes. It is gnuplot 6.0 (patchlevel 2) compiled to WebAssembly, so any script you write here also runs in desktop gnuplot. The downloaded .gp file includes ready-made terminal lines for pdfcairo, cairolatex and pngcairo.
Is my data uploaded anywhere?
No. gnuplot runs in your browser, and your data and scripts are kept only in your browser’s storage on your device.
What is free and what needs Pro?
Making graphs, curve fitting, PNG at 150 dpi, LaTeX (pict2e), the script and the project zip are free. DailyFirst Pro adds 300, 600 and 1000 dpi PNG, vector PDF with embedded fonts, SVG, EPS and epslatex.
What are the disadvantages of gnuplot for research papers?
gnuplot is fast, stable and fully scriptable, but it has limits worth knowing. It has no data frames, so merging, grouping or reshaping data is usually done first in Python, R or awk. Its statistics are basic: fit and stats cover least-squares fitting and summaries, but not ANOVA, mixed models or bootstrap confidence bands. Multi-panel layouts with aligned axes take manual margin tuning, the default style needs a style file before it looks journal-ready, and some syntax changed between versions 4 and 5, so older examples can break. The LaTeX terminals (epslatex, cairolatex) also write two files, so the usual approach for version control is to commit only the script and data and rebuild figures with make, which works equally well with Matplotlib or R.
Which LaTeX output should I use?
Use pict2e when you want one self-contained .tex file whose labels use your document’s font: add \usepackage{pict2e,xcolor} and \input the file. Use epslatex when labels contain LaTeX maths such as $\alpha$: it gives figure.tex plus figure.eps (with pdflatex, also load epstopdf).
More free tools: R & Python Research Lab · Scientific Diagram Maker · Citation Generator