Speed benchmarks of GPT-4 Turbo and gpt-3.5-turbo-1106

OpenAI just released new versions of GPT-3.5 and GPT-4, and there’s a lot of interest about their capabilities and performance. With that in mind, I’ve been benchmarking the new models.

These benchmark results focus on measuring the speed of the new models, and how they compare to previous versions.

Some quick takeaways:

- GPT-3.5 got 6-11x faster.

- GPT-4 Turbo is 4-5x faster.

- The old March (0301) version of GPT-3.5 is actually faster than the June (0613) version. This was a surprising discovery.

More details and graphs in the linked page.