> cheaper than the cost of local electricity only.

Can you explain what that means?

I think they mean that the DeepSeek API charges are less than it would cost for the electricity to run a local model.

Local model enthusiasts often assume that running locally is more energy efficient than running in a data center, but fail to take the economies of scale into account.

▲

croes 4 hours ago | parent | next [-]

Local enthusiasts don’t have to fear account banning.

▲

littlestymaar 6 hours ago | parent | prev | next [-]

I guess it mostly comes from using the model with batch-size = 1 locally, vs high batch size in a DC, since GPU consumption don't grow that much with batch size.

Note that while a local chatbot user will mostly be using batch-size = 1, it's not going to be true if they are running an agentic framework, so the gap is going to narrow or even reverse.

	▲	eru an hour ago \| parent [-]
		Well, different parts of the world also have different electricity prices.

▲

jacquesm 6 hours ago | parent | prev [-]

Some of those local model enthusiasts can actually afford solar panels.

▲

jLaForest 5 hours ago | parent [-]

You are still incurring a cost if you use the electricity instead of selling it back to the grid

	▲	Kodiack 5 hours ago \| parent \| next [-]
		The extent of that heavily depends on where you are. Where I live in NZ, the grid export rates are very low while the import rates are very high. Our peak import rate is 3x higher than our solar export rate. In other words, we’d need to sell 3 kWh hours of energy to offset the cost of using 1 kWh at peak. We’re currently in the process of accepting a quote for home batteries. The rates here highly incentivise maximising self-use.
	▲	dmichulke 4 hours ago \| parent \| prev [-]
		Luxembourg: Purchase price = 2 x sales price, mostly due to grid costs. And this is with no income tax or VAT on sold electricity.

▲

atoav 7 hours ago | parent | prev | next [-]

It means that the electricity you would have to pay if you did the computations yourself would be more expensive than paying them to do it. Part of thst has to do with the fact that China has cheap electricity, also due to their massive push into renewables. Part of that is just economies of scale. A big server farm can run more efficiently than your PC on average.

	▲	AuthAuth 5 hours ago \| parent [-]
		cheap electric due to their massive push on non renewables. There has been no change in the price of electricity during the renewable shift.

▲

jojobas 8 hours ago | parent | prev [-]

China has cheap electricity.

	▲	ericd 8 hours ago \| parent \| next [-]
		Well, also, LLM servers get much more efficient with request queue depth >1 - tokens per second per gpu are massively higher with 100 concurrents than 1 on eg vllm.
	▲	DeathArrow 41 minutes ago \| parent \| prev [-]
		Yes, but the hardware they use for inference like Huawei Ascend 910C is less efficient than Nvidia H100 used in US due to the difference in the process node.