Breaking Limit vs MLX Benchmarks

Breaking Limit is a true cross-platform graphics benchmark for mobile and desktop platforms, enabling effortless comparison of graphics performance across various devices by utilizing the same workload. It leverages cutting-edge Raytracing technology, including Raytraced Shadows and Reflections. The results obtained by running the Official benchmarks are uploaded to Powerboard online scoreboard. There, you may compare your results against others and see where you rank. Included Tests: - Official: The main benchmark that uses standard settings. - Official Native: Uses the same content and settings as Official, but uses device’s display resolution as the rendering resolution. - Custom: Uses the same content as Official, but allows users to change settings from the default ones used in Official.

MLX Benchmark helps you understand how large language models perform on your hardware. Select a model and quantization, set context length and iterations, and run repeatable benchmarks that capture load time, tokenization time, time to first token, throughput, memory usage, and jitter. View detailed metrics grids, visualize comparisons in charts, and manage downloaded model repos directly in the app. Built for developers and AI enthusiasts who want clear, on-device performance insights without leaving their device. Features: - Model selection with quantization details - Configurable prompt, context length, and iterations - Metrics for load, tokenization, first-token latency, throughput, jitter, and memory - Charts to compare runs and devices - Downloaded model management (list, delete, refresh) - Optional Supabase upload for aggregating benchmark results