Building Explainers - DeepSeek Inference
Made another little explainer for myself.
DeepSeek says margins of 70-80% on their inference framework, gives you a sense of how good they are at infra.1
The work behind the next token
Been trying to refine different explainers, made one for myself on this. Comes out ok, obviously it’s a more personal explainer in the style I prefer, but I think it’s not bad for three passes.
Source: deepseek-ai/DeepInfer
How it was made
This time I tried to gather inspiration from around the web for different classic explainers, so influence from Bret Victor or 3Blue1Brown. There are a few other pieces involved in there.
I think the key to it was to have multiple passes to improve the language. There are a lot of tics for the language models that really annoy me. I have a repo where I keep a collection of skills that I use, and I’ve updated that with some of the pieces I’ve removed recently. These are LLM-generated release notes of the updates, but they’re correct.
Related Thoughts
- Building Explainers Random Thoughts - Drone 101: the first explainer experiment, also drawing on Bret Victor’s explorable explanations.
- Cheaper Code Should Mean More Experiments: why it’s worth trying a narrow idea, seeing what works, and trying another version.
Footnotes
-
DeepSeek’s own published figure is $38 of revenue per accelerator-hour at off-peak prices, released alongside DeepInfer on 20 October 2026. The margin is an estimate built on that figure: Latent Space puts it at about 80%, assuming $5 to $8 of all-in cost per card-hour, and other estimates range from roughly 74% to 96%. These leave out cache storage, networking, staff, idle capacity, and training. ↩