Features
Generates optimized CUDA/Triton kernels from any model
32 parallel Coder+Judge agents with MAP-Elites evolution
Pattern RAG with 1,711 CUTLASS and 113 Triton patterns
Up to 5× speedup over torch.compile(max-autotune) on Llama-3.1-8B
100% numerical correctness verification
Automatic Tensor Core optimization (WMMA, TMA) for Hopper
Three optimization modes: --turbo, default, --quality
Dual output: Triton Python kernels or native CUDA C++
Interactive CLI with wizard and benchmark browser
Session management for tracking past optimizations
Supports HuggingFace IDs, KernelBench tasks, or custom files
Credit system: 1 credit per kernel, 1-2 for HF models
Powered by fine-tuned Nemotron 3 Nano 30B at 250K tok/s
Drop-in replacement: same API, zero code changes
Generate Next.js boilerplate with custom content
AI content generation for sections
300+ components and usage examples
37+ landing page components
One-click deploy to Vercel or Netlify
SEO-optimized MDX blog with search, tags, pagination
Dark mode and responsive design
Icon library and chart/data viz components
Pricing, Terms, and Privacy pages
Customizable logo, theme, branding, favicon
Font customization (16 handpicked pairings)
Zip download of generated code
Automatic sitemap and RSS feed
