[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.]

![karpathy Avatar](https://lunarcrush.com/gi/w:24/cr:twitter::33836629.png) Andrej Karpathy [@karpathy](/creator/twitter/karpathy) on x 1.4M followers
Created: 2025-05-27 23:26:44 UTC

So so so cool. Llama 1B batch one inference in one single CUDA kernel, deleting synchronization boundaries imposed by breaking the computation into a series of kernels called in sequence. The *optimal* orchestration of compute and memory is only achievable in this way.


XXXXXXX engagements

![Engagements Line Chart](https://lunarcrush.com/gi/w:600/p:tweet::1927506788527591853/c:line.svg)

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

[Post Link](https://x.com/karpathy/status/1927506788527591853)