Nano-Vllm: Lightweight vLLM implementation built from scratch | Hacker News Reader

Nano-Vllm: Lightweight vLLM implementation built from scratch

github.com

125 pointssimonpure16 comments

Little sparse on the documentation side can't tell at a glance if there is a 1:1 hyperperameter tuneability or if this is an opinionated single path locked soft fpga eval-hacking kind of thing.

EDIT: -- Ok, it's legit, here is an example of it put to use by the makers of the Dolphin OpenSource series of FineTunes:

> Here I implement in nano-vllm, efficient sample-K logit extraction, as described in "Sparse Logit Sampling: Accelerating Knowledge Distillation in LLMs" by Anshumann et. al. Sampling occurs on the GPU, the non-sampled logits do not get copied out of GPU space. I tried to implement this in @vllm_project, but it was a bit too heavy for me to figure out.

https://github.com/GeeeekExplorer/nano-vllm/pull/34

This is an incredible achievement for a solo developer. The dev is from the Deepseek team by the way.

That is crazy! This is so cool ngl.

After seeing the Docker image for vllm jump +5Gb (to 10Gb!) over the past five months, I grew suspicious of vllm's development practices [1]. It's not easy, for sure, to deal with all those flaky python modules [2].

But having the CUDA packages four times in different layers is questionable! [3]

Yet again, as a college mate of mine used to say, "Don't change it. It works."

[1]: https://hub.docker.com/r/vllm/vllm-openai/tags

[2]: https://github.com/vllm-project/vllm/issues/13306

[3]: These kinds of workarounds tend to end up accumulating and never get reviewed back:

- https://github.com/vllm-project/vllm/commit/b07d741661570ef1...

- https://github.com/vllm-project/vllm/commit/68d37809b9b52f4d... (this one in particular probably accounts for +3Gb)

Meta: the Title Casing in the title is pretty obnoxious, "Vllm" is exactly the inverse, casing-wise, of how the project spells its name.

Fwiw op has a small window of time to correct the casing after posting

Love this project, we need more simplifications like this in the current ML environment

Will this end up getting an open ai compatible web server or is that out of scope.

Did anyone else click in excitedly after misreading ‘Vllm’ as ‘LLVM?’

So... It's a language model..? As in, not "large"? I'm a bit unsure of the magnitudes here, but surely "nano" and "large" cancel out

No, vLLM is a thing for serving language models: https://github.com/vllm-project/vllm

Is it more like llama.cpp then? I don't have access to the good hardware.

llama.cpp is optimized to serve one request at a time.

vllm is optimized to serve many requests at one time.

If you were to fine tune a model and wanted to serve it to many users, you would use vllm, not llama.cpp

Here's a super relevant comment from another post https://news.ycombinator.com/item?id=44366418

Appreciate it!

how did u do it? which portion of vllm refactoring allowed u to get such gains.

Reply on news.ycombinator.com