You absolutely can; they are not banning anymore. The bigger problem is that subscription versions of the models are way crappier than when the "same" model is hit via API (Bedrock/Vertex)
You can also make it not count against extra usage.
OpenCode docs show it because Anthropic specifically ambushed them with a PR to remove support so simpletons can't use it easily.
There are many fabs using "last year's" fab processes for defense purposes. Wouldn't be surprised if they had quick and easy way to set up fabs for chip weapons production in bombed out buildings and warehouses. Defense doesn't need civilian fabs. In the end, we, the civilians, stand to lose tremendously.
They will rapidly reindustrialize when the first shots are fired. The EU's goal is the strategic defeat of Russia. What the common people think or want is irrelevant. All environmental and climate legislation that gets in the way will be waivered indefinitely until the war is over 5+ years of drone warfare and 100s of thousands dead.
If you are running an org you should be putting all your private services behind netbird or tailscale at minimum. Zero public infra exposure beyond them.
The article has failed to prove that anybody has taken the bait and left.
> For the researcher, the qualities must, from an international perspective, far exceed what is customary within the international peer group. The institution receives a maximum of €1 million per researcher for the next five years.
Let's be generous and assume you are one of the chosen ones. Your institution will take 20% off the top leaving with you 1million×.80/5 or 160k EUR per year.
After income taxes, your take home pay is €90,868.00 or $103k USD. Not bad for the average man, but not good for a top researcher like they want.
EUR 160k works out to about $182,640. For that level of income in a top tier institution in a state with an income tax like Johns Hopkins in Baltimore, MD you would take home $121,565, or 15% more.
It's the industrial revolution all over again. The basket weavers are clutching their pearls because the new mechanized products are too much for them to understand plus the new machines require complicated operating procedures and ways of doing things
That's why you have to reframe in terms of total cost per task and factor in model token generation quantity and multiply that by the base cost of the model. Then factor in your time value if you dare. Then you should get a more meaningful business metric.
When assessing probabilistic models the plots should be showing the mean a̶n̶d̶ ̶s̶t̶d̶e̶v̶ of many monte carlo simulations not just one line per model and claiming "look this model is more gooder!"
They form these groups to network and do favors for each other and provide cover for each other. After a while you are asked to do increasingly compromising things to yourself to do deeper into/stay part of the club such as blood rituals. Bohemian Grove high priest Henry Kissinger led simulated animal sacrifice rituals attended by many others in the "inner circles" of the top strata of decision makers and wallet holders. They do this because engaging in a common taboo unites the group extremely strongly, and these behaviors are the ultimate taboo.