_hugerobots_

Like the other commenter, I'm confused about the 'just not coding' conclusion. I'm using Qwen 27B on a 5090 at > 100tk/s with 150k context (which isn't enough admittedly), and DeepSeek v4 Flash with 1million context on a gb10/spark. Both of which are performing surface level, and deep needle precision infrastructure architecture. They code 24-7, stupendously.

So if it isn't a comparative ability, now it's a power cost issue? This reads like goal post moving.

Defining productivity is a use-case scenario, and a wildly generalized assumption for most people in this argument. Local infrastructure doesn't need to be sota for absolutely every single need for a dev lab, but it absolutely can be delivered with non-api frontier class models.

Speed vs task-completion is a new conversation and a great point. Whereas the cost to compute doesn't exist in a vacuum, making mistakes costs less, is easier to maintain with granularity and a whole host of other factors when you own the lab.

Local models can be widely used as productive assets. Yes the infrastructure of SOTA API models is engineered specifically for you to be that utility, but the blanket statement that local isn't up to par is intensely short sighted. Billions of tokens per month on local pays for the hardware when compared to sota costs per month.

This 1000%. Data centres don't equate to medium sized labs and businesses. A stack of Macs is up and running without digging trenches, an electrician on staff and a department of PhDs to justify the spend.

Have a 5090, and yes it's very fast. But it's like the worst ADHD team member and requires constant supervision and review from larger models. It's context size on-card is good for super, suuuuuper shallow precision work. The gb10/spark on top of it, that thing can refactor enormous monorepo architecture. The time it takes the 5090 to compact, reiterate and execute a plan is often the same time as the gb10.

It took a very unrealistic amount of time to realize that the IP theft, artificial blockers and corruption had started a lot earlier than presumed. The Nortel dismantling should be taught as an example of infoSec espionage on the levels of corporate terrorism.

Meeting expats from any nation will hold a bias untoward the place they're from, so you're asking a poisoned well how thirsty it is.

You're right agents always short-change the no-paper-crew around here, but I always thought this was just a Vancouver thing. My contracts are in the f50 range for mag7 teams.

Just last night an analyst came into my little startup/store. Asked him to take a look at my latest scratch build. Told me to just get a degree from a cert-farm and go for Bay-Area-level work. But, maybe I need therapy instead, because I just don't know a single colleague at the F50s in this city that say the same thing.

This would be a really nice product to start ups outside the US tech belt. Hubris of treading water in longterm a-series SUs elsewhere, this could be a viable solution if accessible.

Hub and spoke documentation in planning has been absolutely essential for the way my planning was before, and it's pretty cool seeing it work so well for planning mode to build scaffolds and routing.