Compare DejaVu vs. oMLX

DejaVu is a specialized app designed for Mac users that functions as both a private screen-memory and desktop-search tool. When activated, it captures and compresses screen frames whenever there is a change, enabling it to extract and store the text displayed on the device, so you can easily locate previously viewed information by searching for specific words, numbers, or phrases. The search results will provide details about the app, the window, and the timestamp, while a timeline feature allows you to navigate through your activity by the hour, and the My Day section gives an overview of your usage categorized by application and project. The screen memory is securely stored on your Mac, encrypted for safety, and requires Touch ID for access. Users have the option to pause the capture process, exclude specific applications and websites, and export their memory in widely-used formats. The free version retains memory for three days on the main display, while the Pro version offers unlimited memory across all displays, browser extensions, the ability to export data, smart collections, and a unique feature called Ask Your Memory, which provides answers directly on the device. DejaVu is compatible with macOS 26 Tahoe or later and works on both Apple silicon and Intel Macs, while the functionality for generating written AI responses necessitates the use of Apple Intelligence available on Apple silicon devices. This ensures users have a comprehensive and efficient method for managing their screen activity and retrieving information seamlessly.

oMLX is an MLX server specifically designed for macOS, enhancing the efficiency and speed of local AI operations on Apple Silicon. It caters to the functional dynamics of coding agents by implementing paged SSD KV caching, which enables the persistence of cache blocks on disk; this means that previously accessed prefixes can be retrieved quickly across different requests and even after server restarts, thereby eliminating the need to recompute them from scratch. As a result, the time taken to generate the first token in lengthy contexts can be significantly reduced, dropping from a range of 30 to 90 seconds down to less than five seconds after the initial interaction. The server adeptly manages simultaneous requests through a continuous batching mechanism via mlx-lm’s BatchGenerator, which enhances overall generation throughput without requiring requests to queue up behind a single task. oMLX is capable of simultaneously serving a variety of models, including LLMs, vision-language models, embedding models, and rerankers, utilizing LRU eviction to manage memory constraints effectively. Furthermore, it is compatible with any MLX-format model sourced from Hugging Face, such as Qwen, LLaMA, Mistral, Gemma, DeepSeek, MiniMax, and GLM, and can also utilize models that are already present in the standard Hugging Face cache, directories associated with LM Studio, or any custom storage locations, ensuring a versatile user experience. This flexibility in model integration enhances the overall usability and practicality of oMLX for developers and researchers alike.