DeepSeek v4.1 flash runs 23 seconds/token on a 2020 16gb M1 Mac Mini

killingtime74 21 hours ago | parent | next [-]

I did a task that cost about $1.5 on API. It was 4 million tokens. That will take about 2.9 years at this rate.

Estimating an electricity cost in the three digits

a012 a day ago | parent | prev | next [-]

You almost got me, it’s 2.6 tokens/minute

	▲	361994752 21 hours ago \| parent [-]
		I fell into the trap (again)

▲

esperent 16 hours ago | parent | prev | next [-]

That's 23 seconds for 1 token. Unusable, in other words. The title seems to be intentional written to obfuscate that.

▲

spottedmarley a day ago | parent | prev | next [-]

Nice! Totally different subject but, have you tried Ling-mini on it? It's really good for how small it is and you can run it on damn near anything and be useably quick.

▲

gbraad 20 hours ago | parent | prev | next [-]

That is ≈0.043 token p/s. Impressive!

▲

po_westnet26 a day ago | parent | prev | next [-]

[flagged]

▲

BottieZimmie a day ago | parent | prev | next [-]

[flagged]

▲

po_westnet26 a day ago | parent | prev [-]

[dead]