Wondering if anyone is doing performance comparison with gpt4 vs gpt4o. I was doing some tests for summarisation, & question answering. For a parameter like “Time to generate” which is crucial for consumer facing apps, I am not seeing an improvement with gpt4o. Any others with similar issues?
Related topics
Topic | Replies | Views | Activity | |
---|---|---|---|---|
Worse results when using GPT-4o as an evaluator | 2 | 550 | October 1, 2024 | |
Performance Degradation of GPT 4o API | 3 | 352 | October 23, 2024 | |
GPT 4 slow and has timeout - GPT 4 versus 3.5 | 1 | 626 | August 10, 2023 | |
Gpt-4o tokens per second comparable to gpt-3.5-turbo. Data and analysis | 3 | 10395 | August 16, 2024 | |
GPT-4 performance is not acceptable for production use cases | 5 | 2362 | December 21, 2023 |