ChartGPU
An open-source TypeScript charting library that uses your GPU to render 50 million+ data points smoothly.
🔗 Visit ChartGPU
Description
Most charting libraries start to choke once a dataset grows past a few thousand points — the browser has to redraw everything on the CPU, and scrolling or zooming turns into a slideshow. ChartGPU takes a different approach: it hands the actual drawing work to your graphics card, the same way a video game does, so charts with tens of millions of points stay smooth to interact with.
ChartGPU is an open-source TypeScript library that uses WebGPU to render line, area, bar, scatter, pie, candlestick, heatmap, and 3D charts at 60 FPS, even with 50M+ points on screen. It supports real-time streaming data updates through an appendData() API, interactive features like hover tooltips, crosshairs, and zoom, an annotation system for markers and reference lines, and shared GPU device support so multiple charts on the same dashboard don't each spin up their own context. It requires a WebGPU-capable browser (Chrome/Edge 113+, Safari 18+, Firefox with recent versions on Windows/Mac).
💬 Our review
The short version: if you're building a dashboard or financial/scientific tool that needs to visualize genuinely massive datasets in the browser, ChartGPU is a serious, free, open-source option — but if your charts have a few thousand points, a simpler library like Chart.js or Recharts will be easier to integrate and won't require a modern-browser WebGPU dependency.
Against established libraries like Highcharts or D3, ChartGPU's whole reason to exist is scale: those tools generally start struggling well before 1M points, while ChartGPU is built specifically for the 50M+ point case using GPU rendering. The cost of that is WebGPU's still-uneven browser support — it needs a recent Chrome/Edge/Safari/Firefox, so it's not yet a safe universal default the way Canvas- or SVG-based libraries are. For teams whose actual bottleneck is rendering performance on huge time-series or scientific data, that tradeoff is easily worth it; for typical business dashboards, it's probably more library than you need.
💰 Pricing
📊 Global score
🤖 AI-enriched data
100% open source et gratuit, disponible via npm
Pros
Rendu GPU (WebGPU) jusqu'à 50M+ points à 60 FPS
Mises à jour de données en temps réel (streaming)
Multiples types de graphiques dont 3D
100% gratuit et open source
Cons
Nécessite un navigateur compatible WebGPU récent
Overkill pour de petits datasets
Écosystème/communauté plus jeune que D3 ou Highcharts