[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.]

![tphuang Avatar](https://lunarcrush.com/gi/w:24/cr:twitter::43463961.png) tphuang [@tphuang](/creator/twitter/tphuang) on x 24.1K followers
Created: 2025-07-19 13:30:36 UTC

@SiliconFlowAI If we compare the initial Ascend-910 to 910B to 910C. One can see not only higher Tensor computation & core, but more importantly significantly greater HBM & memory speed.
Anyone who has done inference will tell you the latter X are seriously important for large models.

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

XXX engagements

![Engagements Line Chart](https://lunarcrush.com/gi/w:600/p:tweet::1946563324138709322/c:line.svg)

**Related Topics**
[inference](/topic/inference)

[Post Link](https://x.com/tphuang/status/1946563324138709322)