jtatarchuk

jtatarchuk on Hacker News

Re:Ask HN: Why hasn't AMD made a viable CUDA alternative? on Tuesday April 01, 2025 @05:13PM (#43551422)

Ya'll should have come to Beyond CUDA.

https://www.linkedin.com/posts/jtatarchuk_beyondcuda-democra...

Attention is NOT all you need: Qwerky-72B trained using only 8 AMD MI300X GPUs

Posted by jtatarchuk on Tuesday April 01, 2025 @03:31PM
from the substack.recursal.ai dept.

https://substack.recursal.ai/p/qwerky-72b-and-32b-training-large

( Read More... | 3 comments | 10 points | on HN )

TensorWave and MK1 partner to unlock fast inference on AMD

Posted by jtatarchuk on Wednesday May 22, 2024 @11:56AM
from the morningstar.com dept.

https://www.morningstar.com/news/business-wire/20240521368988/tensorwave-partners-with-neuralink-engineers-new-startup-mk1-to-bring-lightning-fast-ai-inference-to-amd-cloud

( Read More... | no comments | 2 points | on HN )

Re:Take AMD MI300X for a test drive on Tuesday April 02, 2024 @12:52PM (#39907958)

Free to try it out and then $1/hr to reserve it.

Re:Take AMD MI300X for a test drive on Tuesday April 02, 2024 @11:54AM (#39907198)

Need compute? Try it free. Limited time $1/hr for the first 3 months.

AMD MI300X vs H100 2.4X Higher Memory Capacity 1.6X Higher Memory Bandwidth 1.3X FP8 TFLOPS 1.3X FP16 TFLOPS

Up To 20% Faster Vs H100 (Llama 2 70B) In 1v1 Comparison Up To 20% Faster Vs H100 (FlashAttention 2) in 1v1 Comparison Up To 40% Faster Vs H100 (Llama 2 70B) in 8v8 Server Up To 60% Faster Vs H100 (Bloom 176B) In 8v8 Server

TensorWave

Take AMD MI300X for a test drive

Posted by jtatarchuk on Tuesday April 02, 2024 @11:54AM
from the tensorwave.com dept.

https://tensorwave.com/

( Read More... | 3 comments | 3 points | on HN )

Re:I Am Sam Altman, President of Y Combinator. AMA on Friday March 18, 2016 @10:30PM (#11316961)

My cofounders and I believe CRM is broken... With a heavy emphasis on management and not the customer... What other areas of CRM do you see need improving?