#Gpt4 unusable
1 messages · Page 1 of 1 (latest)
Hi @raw dagger. I'm kapa.ai, your friendly support bot. I will try to answer your question by searching through the knowledge base. :wave:
Yes, there have been several reports of slower response times with the GPT-4 model. Some users have reported that the average token generation time has increased significantly, making the application almost impossible to use source (https://community.openai.com/t/640146).
The slower response times are somewhat expected as the GPT-4 series models are generally slower than the 3.5T series models source (https://community.openai.com/t/640146).
If you're finding the response times too slow, you might consider using the Dynamic option in the model selector. When this option is selected, OpenAI will automatically choose the best model to use for your request based on a combination of intelligence, capabilities, and speed. In practice, this means that GPT-4 will be used for most tasks. For easier ones, GPT-3.5 might be used to speed up responses source (https://help.openai.com/en/articles/9155758-what-is-the-chatgpt-plus-model-selector#h_a4d64a0a25).
However, please note that downgrading to a faster model like 3.5 Turbo might result in lower response quality source (https://community.openai.com/t/640146).
If you're still experiencing issues, it might be worth reaching out to OpenAI support or the OpenAI community for further assistance.
Note: you can ask me a follow up question by @ mentioning me again :speech_balloon:
@raw dagger kapa.ai is still learning and improving, please let us know how it did by reacting below
Yes, put i payd money to use the gpt4 ... 🙂 It is slower but smarter.
But now is the slowest ever I seen in the past 1.5 year ... Practically is unusable ...
and the gpt4o is also slow
- and the gpt4a is also slow
🙂 just telling my story, like gpt4o is dooing his own