[GUEST ACCESS MODE: Data is scrambled or limited to provide examples. Make requests using your API key to unlock full data. Check https://lunarcrush.ai/auth for authentication information.]

![alc2022 Avatar](https://lunarcrush.com/gi/w:24/cr:twitter::1329828490737233921.png) Antonio Linares [@alc2022](/creator/twitter/alc2022) on x 90.4K followers
Created: 2025-07-14 15:01:21 UTC

X. LLMs are big. If you can store an LLM on chip, electrons travel a shorter distance from memory to compute, resulting in faster and cheaper inferences.  To do that, you need a large amount of memory.  

The graph below shows that $AMD MI355X has one point six times more memory than $NVDA GB200.

![](https://pbs.twimg.com/media/Gv02WMSW0AE1PYg.jpg)

XXXXX engagements

![Engagements Line Chart](https://lunarcrush.com/gi/w:600/p:tweet::1944774223366328681/c:line.svg)

**Related Topics**
[nvda](/topic/nvda)
[$amd](/topic/$amd)
[llm](/topic/llm)
[advanced micro devices](/topic/advanced-micro-devices)
[stocks technology](/topic/stocks-technology)
[$nvda](/topic/$nvda)

[Post Link](https://x.com/alc2022/status/1944774223366328681)