Apple announced on December 6 the launch of MLX, an open source framework designed specifically for Apple Silicon. The goal of this framework is to enable AI developers to build, test, use, and optimize in their projects.
Developer Oliver Wehrens recently shared the benchmark results of the MLX framework on Apple's M1 Pro, M2 and M3 chips. , and compared with the Nvidia RTX 4090 graphics card. OpenAI’s speech recognition model Whisper was used in the test.
Wehrens uses the Whisper model for speech transcription and measures the time it takes to process a 10-minute audio file. The results showed that the M1 Pro chip performed slightly worse than the Nvidia GPU, taking 216 seconds to process audio compared to 186 seconds for the 4090.
However, the performance of the new generation of Apple Silicon has been greatly improved. For example, another tester ran the same audio files on an M2 Ultra with 76 GPUs and an M3 Max with 40 GPUs and found that those chips took less time to transcribe than the Nvidia GPU.
There is a significant difference in power consumption between Apple Silicon and Nvidia graphics cards. Specifically, comparing the power consumption of an Nvidia 4090-powered PC between running and standby states, that's an increase of 242 watts. In comparison, the difference in energy consumption between active and standby mode on a MacBook with 16 M1 GPU cores is only 38 watts.
It is worth mentioning that the Nvidia 4090 GPU price starts from $1,599, not including PC, which is different from The 2022 M3 MacBook Pro is the same price! These results highlight Apple's progress in artificial intelligence and machine learning capabilities and may be the beginning of better performance for Apple products. With the MLX framework now open source, it provides developers with access to wider application and innovation.
The above is the detailed content of Beyond Nvidia RTX 4090: M3 Pro AI achieves huge breakthrough in running scores. For more information, please follow other related articles on the PHP Chinese website!