Anfield Stadium.

Raining in Old Trafford.

Etihad Stadium.

Tottenham Hotspur Stadium.

As a follower and active contributor for inference platform, I created the llmaz project to provide an unified inference platform for LLMs and also joined the AIBrix community to build the next-gen GenAI infrastructure.

I just read the article about the business mode with open source commercialization in fit2cloud, which is really impressive, so I wrote down some highlights here for future reference. Note that, the original article was written in Chinese.

Due to DeepSeek-V3 technical report, it says:

In addition, both dispatching and combining kernels overlap with the computation stream,
so we also consider their impact on other SM computation kernels.
Specifically, we employ customized PTX (Parallel Thread Execution) instructions and
auto-tune the communication chunk size, which significantly reduces the use of the
L2 cache and the interference to other SMs.

then people are saying like DeepSeek is breaking the Nvidia core moat - CUDA by employing the PTX directly, but is that true?

Feed Nania with bananas.

A month ago, Ilya Sutskever, the ex co-founder and chief scientist at OpenAI, gave a talk at the NeurIPS 2024 and announced that: PreTraining is Over. He reveals the fact that the available data of internet for training large language models is exhausted, which somehow challenges the scaling law (for short, the performance and accuracy of AI model improves as a function of increasing the scale in model size, dataset size and compute power).

Days ago, somebody asked me why I want to contribute to open source, what do I expect from the involvements, I told that it’s a nature motivation as an engineer, it’s true, but I want to elaborate more here and write down my understandings.