@PandaTalk8: These test results are stunning. The original poster tested the DS4 inference engine written in C by @antirez, and local deployment seems incredibly fast. The good news is that only 128GB of RAM is needed to run a local model equivalent to GPT-4o. The bad news is that you need a MacBook Pro with 128GB of RAM.
Summary
This article reports on tests of the DS4 inference engine written in C by @antirez, noting its impressive speed when running a GPT-4o-equivalent model on a MacBook Pro with 128GB of RAM.
View Cached Full Text
Cached at: 05/10/26, 02:28 PM
The results of this test are absolutely stunning.
The original poster tested the DS4 inference engine recently written in C by @antirez. The local deployment looks incredibly fast.
The good news is that you can run a local model equivalent to GPT-4o with just 128GB of RAM. The bad news is that it requires a 128GB MacBook Pro.
@XianyuLi, your 128GB MBP purchase was not wasted. I was wrong—I’m going to work hard to upgrade my computer.
Similar Articles
@VincentLogic: Discovered an amazing open-source project! Redis creator antirez made a splash! ds4 — DeepSeek V4 Flash local inference engine, optimized for Mac Metal, topping GitHub charts for days! And here's the killer part: 128GB…
Redis creator antirez released an open-source project called ds4, a DeepSeek V4 Flash local inference engine optimized for Mac Metal, featuring disk KV caching, ultra-long context, and excellent performance.
@ttasanen: Just fired up DS4 by @antirez on my Mac Studio M3 Ultra 256GB and man, it’s seriously impressive. A clean, purpose-buil…
DS4 is a specialized inference engine by antirez designed to run DeepSeek V4 Flash locally on high-end Mac hardware, featuring optimized KV cache handling and 1M context support.
@mylifcc: I'm already running Gemma-4-12b on my Mac. Tech stack: llama.cpp + GGUF Q4_K_M + Metal 32K context, local OpenAI-compatible API. Measured about 36 tok/s, resident RSS about…
User shares their experience using llama.cpp with the GGUF Q4_K_M quantized version of Gemma-4-12b on a Mac, achieving local inference speed of about 36 tok/s and memory usage of about 10GB.
@Michaelzsguo: So you bought the 128GB MacBook Pro. Now the question is not, “Which local model gets the highest TPS?” It is: which se…
This thread recommends a local AI coding stack for the 128GB MacBook Pro, using Qwen 3.6 model with MLX server and specific configurations for reliable coding assistance.
@sitinme: There's a pretty interesting open-source project called Cider, specifically designed to accelerate local AI inference on Macs with Apple Silicon chips. Many people buy a Mac mini or MacBook Pro and want to run models locally, but often encounter issues like insufficient speed and high memory usage. Actually...
Cider is an open-source project designed for Apple Silicon Macs, accelerating local AI inference by fully leveraging the computing power of M-series chips. It is compatible with the MLX ecosystem, supports models like Qwen and Llama, and is easy to install.