News for hackers. Stuff that matters.

llama-mini

llama-mini on Hacker News

Re:R1 Computer Use on Thursday February 06, 2025 @03:46PM (#42966315)
It seems a placeholder for now? No content? Right?
Re:Distributed Transactions at Scale in Amazon DynamoDB on Wednesday January 15, 2025 @03:01AM (#42708502)
> it only works well for well defined use-cases.

Sounds to me DynamoDB works well for well defined use-cases. That to me is a plus!

Re:Shockley Semiconductor Laboratory on Tuesday December 24, 2024 @07:21PM (#42505959)
No, the man's personality is the engine of his achievement.

It's just that ordinary human do not really rewards genius.

For the man being different, he probably can be Gorden Moore, who just noticed a pattern, and then become monumental figure.

Populous rewards popularity, not genius. And being ignored by them is not a failure, it's a necessity of greatness.

Re:Shockley Semiconductor Laboratory on Tuesday December 24, 2024 @07:18PM (#42505951)
https://substackcdn.com/image/fetch/f_auto,q_auto:good,fl_pr...

Look at the picture, you can notice who's shockley by just observing their facial expression. The cold-and-smart-than-you shockley in the middle. Shockley's personality becomes his mortal enemy, but that personality is also ncessary for his achievement.

If strength cannot become your weakness, than it's just mediocrity.

Re:Making AMD GPUs competitive for LLM inference on Tuesday December 24, 2024 @12:04AM (#42499843)
From Lamini, we have a private AMD GPU cluster, ready to serve any one who want to try MI300x or MI250 with inference and tuning.

We just onboarded a customer to move from openai API to on-prem solution, currently evaluating MI300x for inference.

Email me at my profile email.