this is a failure mode of consumer co-op that never occured to me. I thought the idea was bullet proof.

consumer co ops are broken because you can pay to get in. any bad actor whos rich enough can get thousands of people to join and vote in a corrupt way. and if you make it invite only you defeat the whole point.

imo the best way to fix is with a hybrid model where consumers pay member fees to get discounts but decision making is down to workers, basically how it was before shareholder supremacy.

>Estimated bill updates remain paused

Wait what if someones actually getting usage spiked

Hackers rejoice!

If you owe AWS 140B dollars its their problem ;)

Something like this together with MCP can replace APIs for 3rd party integrations. You just give it instructions to "post a message in slack" and provide it slack MCP tools and it figures out the rest on its own. No need to read up on slack API docs or worry about breaking changes.

Point taken but there isnt any western providers there yet. Power is cheaper in china.

These models are open and there are tons of western providers offering it at comparable rates.

As this is a new arch with tons of optimisations, it'll take some time for inference engines to support it properly, and we'll see more 3rd party providers offer it. Once that settles we'll have a median price for an optimised 1.6T model, and can "guesstimate" from there what the big labs can reasonably serve for the same price. But yeah, it's been said for a while that big labs are ok on API costs. The only unknown is if subscriptions were profitable or not. They've all been reducing the limits lately it seems.

Is there evidence that frontier models at anthropic, openai or google or whatnot are not using comparable optimizations to draw down their coats and that their markup is just higher because they can?

not soooo much though. It's heavily subsidized for residential consumption, but industrial power rates are almost comparable to the US (depends on the state you go to etc).

This looks like the same suicide inducing type of crap by google that previously only android devs on playstore were subject to.

I'm permanently banned from the Play Store because 10+ years ago I made a third-party Omegle client, called it Yo-megle (neither Omegle nor Yo-megle still exist now), got a bunch of downloads and good ratings, then about 2 years later got a message from Google saying I was banned for violating trademark law. No actual legal action, just a message from Google. I suppose I'm lucky they didn't delete my entire Google account.

llm tool call -> mcp client validates the schema -> mcp client calls the tool -> mcp server validates the schema -> mcp server responds with the result -> mcp client passes the tool result into llm

not a developer.

what happens if this schema validation fails here - what will the mcp server respond with and what will the llm do next (in a deterministic sense)?

llm tool call -> mcp client validates the schema -> mcp client calls the tool -> mcp server validates the schema

They often do fail, at the client level you can just feed the schema validation error message back into the LLM and it corrects itself most of the time. If not the LLM throws itself into a loop until its caller times it out and it sends an error message back to the user.

At the server level it's just a good old JSON API at this point, and the server would send the usual error message it would send out to anyone.