Infinite typing
bot no longer follows through with response after updating agent instructions to match Gemma 4 from Gemini 2.5.
nevermind it followed through, but the response time was over 4 minutes
Yes, it's a bit long, but we should expect it to be on par with the speed of Gemini very soon. There's a new release from Google, and we are expecting the inference provider on OpenRouter to pick up soon.
BTW, for the premium, we are going to launch a couple new sub-features, so you can choose one of the premium models (going back to Gemini if you want - but note that if there are failure from google, we can't do anything to provent it, and the main reason we switch is because gemini was not usable for like 20 days).
Adding llm evaluation is another one.
And finlly, we are gonna finally have a ticket add on version for Spark
and yes, very awre of the speed issues. we are gonna keep an eye on it for now, and if nothing changes in a next 2 weeks, we re gonna do something on our end about it
