Primate Labs, a benchmarking company, released Geekbench AI 1.0 on Thursday
To standardize performance ratings across platforms, the application, which is presently accessible on Android, Linux, MacOS, and Windows, applies Geekbench’s principles to machine learning, deep learning, and other AI workloads. It is the successor to Geekbench ML (machine learning), which was introduced in 2021 and is presently at version 0.6.
Primate Labs explains that in recent years, companies have converged on using the term “AI” in marketing these types of workloads. “We believed that it was time for an update to guarantee that all individuals, from performance enthusiasts to engineers, comprehend the purpose and functionality of this benchmark.”
A new iteration of its own AI model benchmark was announced by OpenAI, the manufacturer of ChatGPT, earlier this week. SWE-bench Verified is a “human-validated” offering that employs human validation to assess the effectiveness of models in resolving “real-world issues.”