It's been quite some time that the notifications coming from antigravity are appearing to me as they're from "jetski"? is anyone else experiencing the same issue? is this a bug with AG or is it just me? am I missing something? wtf is jetski?
I've been testing 3.5 flash lite for coding and it surprised me to be nearly or as good as the old 3 flash model but way faster! I am impressed since the older 3.1 flash lite was awful, worse even than gemma 4 but now with this new version the story changes completely.
Now what I don't understand is why is google not incluidng this model for AG to save us token burning with 3.5 flash and/or 3.6 flash? I bet for sure its way less expensive and we could easily use it to do regular work without runnig out of quota the second day of the week...
I was thinking the other day the fact that we still have Opus 4.6 and even though we went from 3 flash to 3.5 and from 3 pro to 3.1 pro, still Opus 4.6 is beating them in every way! its a curious fact and something that makes me wonder if they're actually improving their models or just making them more expensive?
Try it out with your Ollama Cloud subscription and/or your local models!
Try it out with your Zai api key and GLM models!
Hi everyone I recently launched this project that runs natively with Ollama, currently the compiled setup file is only available for windows and the whole project been only tested on windows, even though the project is electron based and virtually could run on any system.
Give it a try it has many features that makes it superior to any LLM web client like Gemini or ChatGPT, it gives your models tools that actually help it be more helpful and smarter like the ability to give you exact calculations without hallucinations regardless of the model. You won't regret trying it out:
