Ten days ago I had an experience with Gemini 3.8 flash that made me wonder if I was being routed to a different model under test. I was trying to use rocm with llama.cpp on my 128gb Strix Halo but could only get it to run Vulkan. I pasted the error message into agy and it proceeded to attach GDB to my GPU driver, reverse-engineer the kernel queue ioctl interface, and author an LD_PRELOAD C shim to get ROCm llama.cpp working on my Strix Halo. My jaw was hanging open the whole time.
Edit to add the fix: <a href="https://gist.github.com/birep/6f2c8d490c7a29820997d57bd654c351" rel="nofollow">https://gist.github.com/birep/6f2c8d490c7a29820997d57bd654c3...
Can confirm - I am HEAVY claude user, but always like to check with AGY and CODEX in between. AGY with Gemini 3.8 flash cooked last couple of times and CODEX is basically out of the mix for me
I find Opus 5.5 is better at giving a high level adversarial "should you do this to begin with" where GPT-6.1 is happy to go down any wrong path.
Also canceled my ChatGPT Pro $200/mo subscription. Their Oct 30 price hikes and slow GPT-6.1 model has me looking for alternatives.
taylorfinley · · focus · HN ↗
Edit to add the fix: <a href="https://gist.github.com/birep/6f2c8d490c7a29820997d57bd654c351" rel="nofollow">https://gist.github.com/birep/6f2c8d490c7a29820997d57bd654c3...
amanguliani · · focus · HN ↗
onlyrealcuzzo · · focus · HN ↗
I'm using it to run overnight tasks, and that's it until my quota runs out.
Canceled my subscription.
unconscionable · · focus · HN ↗
Also canceled my ChatGPT Pro $200/mo subscription. Their Oct 30 price hikes and slow GPT-6.1 model has me looking for alternatives.
directdev · · focus · HN ↗
[dead]