I am trying to benchmark it now, but:
- It doesn't seem available in EU (?)
- Using a VPN seems to sort of fix it, but it's way slower than I expected, when everyone was praising it, it feels like the speed is slowly ramping up
- Cost is $2/$6 for <200k context only, above that, cost is $4/$12
- GLM-5.2 still seems smarter, faster and much cheaper:
https://aibenchy.com/compare/x-ai-grok-4-5-medium/z-ai-glm-5...
I concur that GLM-5.2 still seems "better" in my experimentation I did with it tonight, although Grok 4.5 is cheap if you can tolerate the way a Cursor subscription works.
Instead of a VPN, might I suggest running litellm proxy on a server in the US and connecting to that
What would the advantage be?
You have control over how the connections work and what they do, rather than just relying on the VPN exit server routing to get you to the right place.
With things like a fly machine or whatever, you might even be able to run it for free