Comfy Portal vs MLX Benchmarks

Comfy Portal puts the power of AI image generation in your pocket. Connect to and control your ComfyUI servers - whether running on your home computer or cloud services like RunPod - all from your iPhone. [Key Features] Flexible Server Management • Connect to multiple ComfyUI servers seamlessly • Support both local network (home computer) and cloud-based servers • Easy server addition and switching • Monitor server status in real-time Remote Generation Control • Control image generation through your ComfyUI server • Adjust generation parameters on the go • Select and switch between different base models Remote Access & Control • Control your ComfyUI instance from anywhere • Real-time generation monitoring • View results instantly on your device • Save and share generated images directly Mobile-Optimized Experience • Intuitive touch interface for parameter adjustment • Clean and efficient model selection • Easy server management dashboard • Responsive design for all screen sizes [System Requirements] • Requires iOS 15.0 or later • Requires access to a ComfyUI server (local or cloud-based) [Note] This app serves as a remote controller for ComfyUI servers. You'll need an active ComfyUI instance running either on your local network or a cloud service to generate images. Privacy Policy: https://shunl12324.github.io/comfy-portal/privacy/ Terms of Service: https://shunl12324.github.io/comfy-portal/terms/

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