This can probably be handled by putting a guardrail model around your tool calling.
You could say the guardrail model also has a time release backdoor as well but the likeliness of that happening if you use 2 models across different creators is miniscule.
I honestly prefer this as opposed to what Anthropic has done in the past which is to continue accepting users even if they don't have the capacity to serve and behind the scenes tone down everyone's limits.
In time, what do you think is gonna happen with the Google Gemini models, I predict that in time Apple will just dump it when they’re ready… No different than dumping Intel. In similar Qualcomm is probably coming up in the near future 2027-2028.
> If market followed rational logic, you could pre-calculate which stocks will be what price and never lose money.
The fundamentals are unpredictable, so even a perfectly rational planning (suppose such thing could exist) would lose money sometimes. Not in the long run, but long run doesn't matter if a single wrecked ship can wreck you.
You're just being pedantic. To a sufficient advanced being (eg God), the whole universe is rational/calculable/etc. Are you the type who says "you're wrong! It's 9:01" when someone says it's 9?
I have used local models (around 128 gb) and the big proprietary models, and while I do want local models to win, it's important we keep the expectations of local models realistic. There are many blog posts about how local models today can fully replace some of the proprietary models and in some cases its true for the much smaller proprietary models, its very clearly much more behind the larger models.
You can be far more ambiguous with your tasks with the larger proprietary models as opposed to the local models. You can achieve the similar results with local models but you need to be much more detailed in your prompt.
One of the biggest things about running these local models is that the harness matters almost just as much as the model too. Codex is optimized for GPT models, CC is optimized for Claude, Cursor has a great harness that works very well across these providers. It took me a couple of iterations of the different harnesses to find one that would work well with the smaller Qwen models to do local coding.
So because of threats to cancel their claude subscriptions and outrage from the community about the invisible guardrails, only then they decided to walk back their stance?
Seems like they would've kept the invisible guardrails if it didn't hurt their bottom line.
> So because of threats to cancel their claude subscriptions and outrage from the community about the invisible guardrails, only then they decided to walk back their stance?
The possibility that the news about "fixing" the "overly aggressive" nerfing of the tool will drown out news about how mismatched the hype and the performance of Mythos and Fable is surely just a bonus.
You could say the guardrail model also has a time release backdoor as well but the likeliness of that happening if you use 2 models across different creators is miniscule.