Hi
I have a problem. I use AI models using openrouter api and other openai compatible APIs (/chat/completions endpoint), but I have a problem with some reasoning models, especially those which takes a lot of time for reasoning (like gpt5-high, deepseek r1, gpt5-pro, o3-pro, ...). the problem is that most of the time when the reasoning takes a lot of time, the answer get caught off and doesn't continue, so I don't see any answer. I want you to see if it's a common issue or not. at first I thought it may be a problem with my internet, so I tested it on a VPS and I got the same issue. then I thought maybe it's an issue the the api provider, so I tested other api providers who gave these models, but again I got this problem.
#Problem with some reasoning models
old elm
@cyan breach
quaint yewBOT
I'm an AI assistant - my answers may be incorrect and I cannot see screenshots. Please verify information with the OpenRouter docs.
To help debug this issue, I need a bit more information:
- What exact error message or response do you receive when the answer is cut off (e.g., timeout, incomplete response, empty message, etc.)?
- What is the exact model identifier you’re using (e.g.,
openai/gpt-5,deepseek-ai/deepseek-r1,openai/gpt-5-pro,openai/o3-pro, etc.)? - Can you share the relevant code snippet or API call you’re using?
These details will help determine if this is a known model/API limitation, a timeout issue, or something else.
-# This is an automated response powered by Answer HQ. Use !mute to stop the bot from responding.
I'm an AI assistant - my answers may be incorrect and I cannot see screenshots. Please verify information with the OpenRouter docs.
To help debug your issue, I need more details:
- What is the exact error message or response you receive when the answer is cut off or does not continue?
- What is the exact model identifier you are using when this happens (e.g.,
gpt5-high,deepseek-r1,o3-pro, etc.)? - Can you share a code snippet of how you are making the API call (including relevant parameters like
max_tokens,timeout, etc.)?
These details will help me (and the mods if needed) pinpoint the exact cause.
-# This is an automated response powered by Answer HQ. Use !mute to stop the bot from responding.
quaint yewBOT
🔇 Thread muted. The bot will no longer respond here.
cyan breach
Hi
I have a problem. I use AI models using openrouter api and other openai compa...
do the requests show up on OpenRouter's activity page?
old elm
do the requests show up on OpenRouter's activity page?
I don't know, I'm using my friends key. here is the cherry studio (the app I use gpt 5 pro) settings.
cyan breach
I don't know, I'm using my friends key. here is the cherry studio (the app I use...
Turn off "set max tokens"
That's why your responses are being cut off
old elm
Turn off "set max tokens"
Did it, but still got the same problem
cyan breach
Did it, but still got the same problem
Use a cheaper model for now so you can save money while testing
Try it with a no thinking model like Mistral Nemo
old elm
Use a cheaper model for now so you can save money while testing
no problem bro. other models like mistral nemo work, even gpt-5-high work in %80 of the time, but this gpt 5 pro only work like %10.
is this a common problem? do other people have also this problem with gpt 5 pro or gpt 5 high?
cyan breach
is this a common problem? do other people have also this problem with gpt 5 pro ...
It's not common
Hmm
You will have to ask your friend to send a screenshot of the activity page so I can see more information