Hacker Newsnew | past | comments | ask | show | jobs | submit | RussianCow's commentslogin

I don't see any kind of input cache discount listed on your pricing page. Do you offer that, or is all input priced the same?

The issue is that all input (including context) counts towards that limit. So 10 requests with 50k of context will blow through the limit, even if little to no output was generated, which is incredibly easy to do with agentic workloads.

This is pretty terrible advice when there are dozens of AI inference providers out there serving great models with significantly more cost effectiveness than you'd get from buying your own hardware.

I don't trust any company with their word on anything. Luckily, privacy policies are legally binding.

> Luckily, privacy policies are legally binding.

Companies violate them all the time and massive leaks happen a lot.

The punishments are trivial.


GPDR fines in the EU are NOT trivial.

> Luckily, privacy policies are legally binding.

Laws are violated all the time. The graveyard is full of people who had the right-of-way at a crosswalk ...


Why does that matter? Legally binding just means "slightly more expensive when we get caught"

Presumably the number that OpenRouter shows is averaged across all requests.

Most providers do what's called "prefix caching", where each turn in a session is cached such that sending new messages with the exact same "prefix" (set of previous messages) gives you the cache read price on that input instead of the full price. As long as you're not changing your system prompt, available tools, etc mid-session, you automatically benefit from this.

But presumably everyone in your company/team is using Jira, so it's not an "ad" because it's a product already used internally. Claude is appending these links to all commits by default, whether or not others on the team use Claude. Those are very different things.

How else would you expect them to calculate it?

Do you really think they're docking points because cache invalidation due to provider switching? Seriously llms are frying ya'lls brain.

They're not "docking points", they're calculating it in the most straightforward way. If I start a session and the majority of requests are sent to Provider A, and my last request gets routed to Provider B, I have a 0% cache hit rate with Provider B. I'm very curious how else you expect this to be calculated? Do you think they're completely omitting requests that switch providers mid-session?

FWIW, I get significantly higher than listed cache hit rates when I pin my session to a specific provider, which is further evidence of the above.


Why wouldn't you only calculate consecutive requests with the same provider....

This is very much NOT my experience in practice, even though it's how I would expect it to work. OpenRouter will happily bounce you between several providers (none of which have downtime) even within the same session. Requesting specific providers is the only way I've been able to hit a cache rate above 90%.

Same experience here. It would choose 2 providers and then bounce between the two every 5 requests or so.

I don't know why there's no "Pick the cheapest provider above nTPS on first request and stick until cache bust" setting.


That explains the difference, but what's the purpose?

The simulator doesn't have a camera. The android emulator allows to configure the camera and use a picture, it's very handy to test your qr/barcode reader without needing to use a physical device. Maybe this also allows to use an image as camera?

It's useful for security research since you can do kernel debugging and inspection of the device not possible in the simulator.

As others pointed out it's also useful for e.g. click/spam farms which need a "real" iPhone.


You can test your application, game, website without a physical phone?

Can do that with the simulator too

Running on ported userspace components, with a shim, isn't the same as running on the OS in itself.

That really matters, sometimes.

But not the vast majority of the time.

(source: been coding iOS since pre-SDK iPhone OS 1.0)


Can you give some examples of where it matters? I'm genuinely curious.

Testing AR apps, or any app that relies on hardware you can't simulate easily.

https://www.youtube.com/watch?v=T43b5ywnYpo


It’s been years but I ran into a problem with a notification extension years ago. Worked fine on simulator, not on device. Turned out it was because I was using a HEIF image and (I think) the on-device extension was trying to use hardware decoding and didn’t have access to it. Meanwhile the simulator had no such sandboxing restrictions.

The difference doesn’t matter until it does, and then it’s infuriating to work out what’s going on.


> The difference doesn’t matter until it does, and then it’s infuriating to work out what’s going on.

I love this quote, reminds me how annoying it can be when something isn’t reproducible locally… only in production…


Last I checked, VPN network extensions wouldn't run in the simulator.

Is the simulator faster than this? I would suspect so.

I have had instances where the app runs fine on the simulator but crashes on the device. This might solve that.

You cannot run Network Extensions on the simulator

If functional, you can download apps from the App Store, sign in to iCloud.. probably do a bunch of agentic stuff as if you had a real device

Looks like apple services compatibility is not currently supported for this [0]

[0] https://github.com/Lakr233/vphone-cli/discussions/175


If you read a bit further, there's a jailbroken variant that has Sileo and Trollstore installed, so you can run whatever IPAs

Where?

Edit: ah now I see, in the main readme under heading Firmware Variants. https://github.com/Lakr233/vphone-cli#firmware-variants


Exactly. It means it cannot download apps from the App Store or use Apple services and Apple knows this deliberate limitation.

Spam. Click farms. Etc.

It would also make testing far easier. I got bit with something that failed on real life iPhones with language set to es_US that worked fine on the simulator.

oh. is this why icloud spam texts are running wild now

No, it’s just trivial to generate iCloud email addresses that won’t get picked up by spam filter.

Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: