there is nothing stopping anyone from canceling. They make an offer, you are free to refuse it
It's the same idea however you characterize it.
And appengine
I hope you can find peace. All the best to you
100% right. Martin Fowler cited Martin Thompson of LMAX who cited Sir Jackie Stewart of Formula One
that's probably the best money this country has ever spent
Off the top of my head, I'm guessing we're missing sparse attention. But I'll run your challenge through and see where the gaps are. I promise I'm telling the truth :)
at 15 tok/s
sure. so is 2x power worth 10x perf? I think it is in most cases.
can confirm. I dont' know why people spend huge money on these and Spark. The 5090 is running qwen 3.8 at 200+ tps!! That's 1-2 orders of magnitude faster.
Til there are like 50 s3 clones
that's boxen in my case and I'm sure others
Meanwhile, this site is covered with ads that are impossible to close
can't second ninfer enough. amazing tech
I've had the opposite experience where we're refactoring the gnarliest shit anyone's ever seen because AI can actually understand it well enough to decompose, test, refactor, etc.
Genuinely curious why. I'm not a Googler and don't follow closely.
It has a new name!
Not sure if it changed but this is written by NYTimes. FWIW I think the WSJ is the best news source available and does not match this description.
patent: https://patents.google.com/patent/US7680791B2
this is the idea of opusplan https://code.claude.com/docs/en/model-config#opusplan-model-...
I posted elsewhere but if you have more space try gemma4 31b
I second unsloth models. I'm using them over blackwell-oriented nvfp4 models as they are (empirically) top quality and performance.
FWIW I'm running gemma4 31b on my 5090 and it's pretty great as well. QAT, MTP, 128k context. I liked Qwen 3.6 27b too, it just seems that Gemma4 is a bit underrated.
Chrome OS Flex has worked out nicely for my mother-in-law's potato
When you leave for the competition
there is nothing stopping anyone from canceling. They make an offer, you are free to refuse it
It's the same idea however you characterize it.
And appengine
I hope you can find peace. All the best to you
100% right. Martin Fowler cited Martin Thompson of LMAX who cited Sir Jackie Stewart of Formula One
that's probably the best money this country has ever spent
Off the top of my head, I'm guessing we're missing sparse attention. But I'll run your challenge through and see where the gaps are. I promise I'm telling the truth :)
at 15 tok/s
sure. so is 2x power worth 10x perf? I think it is in most cases.
can confirm. I dont' know why people spend huge money on these and Spark. The 5090 is running qwen 3.8 at 200+ tps!! That's 1-2 orders of magnitude faster.
Til there are like 50 s3 clones
that's boxen in my case and I'm sure others
Meanwhile, this site is covered with ads that are impossible to close
can't second ninfer enough. amazing tech
I've had the opposite experience where we're refactoring the gnarliest shit anyone's ever seen because AI can actually understand it well enough to decompose, test, refactor, etc.
Genuinely curious why. I'm not a Googler and don't follow closely.
It has a new name!
Not sure if it changed but this is written by NYTimes. FWIW I think the WSJ is the best news source available and does not match this description.
patent: https://patents.google.com/patent/US7680791B2
this is the idea of opusplan https://code.claude.com/docs/en/model-config#opusplan-model-...
I posted elsewhere but if you have more space try gemma4 31b
I second unsloth models. I'm using them over blackwell-oriented nvfp4 models as they are (empirically) top quality and performance.
FWIW I'm running gemma4 31b on my 5090 and it's pretty great as well. QAT, MTP, 128k context. I liked Qwen 3.6 27b too, it just seems that Gemma4 is a bit underrated.
Chrome OS Flex has worked out nicely for my mother-in-law's potato
When you leave for the competition