On our journey at Neuralwatt (we are likely the provider he's using as we are the one that does all the energy observability and reporting in our cloud and I'm the CTO there) we quickly discovered that there is a large disconnect between what is in the press (like the rest of the AI narrative very subject to worst case but possibly unlikley future extrapolation of current trends) and what we see on the ground. We actually spend most of our time focused on various way energy constraints (including maximizing tokens/joule while maintaining perf) manifest in current datacenters and provide better ways to get more tokens out of the energy that are already in these data centers or already available but hard to utilize on the grid. So its really a technical constraint problem <it>today</it> rather than an impact problem and I think the press narrative could be better about this.

Regarding total costs relative to the pure energy costs it is multiple orders of magnitude different but also realize in the datacenter the energy is the pure commodity while almost every other component has huge margins driven by lack of supply. I do think over time this might get closer together (more competition on HW might lower margins) while energy might become more of a bottle neck (raising the energy prices).

> With the talk of AI data centers' impact on the world, you'd think this would be 10x to 100x the amount of energy in order to get the effects they're using here.

This was going so well until this. Everything at scale has environmental impact because you centralize the downside and distribute the upside. This is an important alienation, but it hides the amount of heat, noise and impact on distribution that datacenters have on local infrastructure and environment.

It's running a normal central home heat pump/air conditioner for one hour, or eight hours of playing on a gaming PC. Yeah, the propaganda around datacenter resource usage has massively outrun the truth, which makes me think there are some very interested parties pushing behind the scenes.

I agree but it does worry me how fast my usage is increasing. Two months ago I was using 10x less tokens and probably not much more than 5kWh on inference. This month about 30kWh on inference. If it becomes more affordable, is there going to be another jump? Not quite sure

"the talk of AI data centers' impact on the world" has been wildly exaggerated and you can see here that this is the least impact of any major new technology in the history of industrialization

Hmm I might need to rephrase. Initial challenge was to use GLM 5.3 Flash and I was on the non-Flash version for the whole vibe coded build. Just wasn’t paying attention and I didn’t realize that one session was a quarter of the month’s spend and 150% of the budget (big price difference between models)

Yeah, I clicked expecting a review of GLM 5.3 Flash, but the article was more a retrospective about what he learned during his September challenge: "Only use GLM 5.3 Flash for one month"

He said his experiment was a failure because:

1. He accidentally spent 450M tokens vibe coding with the wrong model, instead of GLM 5.3 Flash.

2. When he used GLM 5.3 Flash, it was sometimes slow. So he switched to other models (Deepseek / Qwen) instead. His guess to why it was slow: GLM 5.3 Flash was so good that the providers were congested.

3. He still needed to use other models besides GLM 5.3 Flash, for R&D and benchmarking.

His takeaways from doing the experiment were:

1. Measure local usage more.

2. Experiment with agent orchestration, with bounded goals.

3. Don't count other models that are used for R&D.

4. Play with Jev.

5. Include experiments with flagship models to compare with cheap open models.

His conclusion about GLM 5.3 Flash: Probably viable for day to day work, but he'll have more thoughts next month.

Some vague commentary about performance with what appears to be assumptions about GPU availability, but no clarity about which inference provider is being used. If ZAI is assumed, I believe they aren't subject to the assumptions in the post based on what they've said publicly, but if they were using some other provider, perhaps.

The second reason appeared to be simply "because we chose not to". The post seems to be pretty much content-less in any practical sense. I clicked on it because I do quite like this models average performance and I was hoping to see some kind of review content.

GLM-5.3-flash is my implementation model after GLM-5.3 writes the plan.

It's an excellent workhorse. When I am running out of my GLM quota I switch GLM-5.3-flash to DS-4.1-flash.