Basemark GPU vs MLX Benchmarks

Basemark® GPU is a multi-platform, multi-API 3D-graphics benchmark. It enables a comparison of the graphics performance of different smartphones and tablets. You can even compare the performance to Notebooks or PCs. The desktop version runs an AAA quality game-like workload by default, but also offers a test identical to the mobile version in this app. Basemark GPU also allows the user to compare their device to others around the world. For that, this free version of the benchmark always submits test scores to the Basemark Power Board web service. If you need a Basemark GPU license for a commercial usage please contact us. To circumvent VSync limitations on mobile devices, we render each benchmark frame off-screen and display only a miniature image of each frame on the screen. This way we can make sure that no frame is dropped, and the results are accurate. If you want to see the graphics in full glory, please choose Experience Mode.

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