My iPhone 16 Pro Max produces garbage output when running MLX LLMs • Hacker News Board

Re: My iPhone 16 Pro Max produces garbage output when running MLX LLMs

#121

Post by watersb » Mon, Feb 02, 2026, 5:46 AM UTC

Buttons840 wrote: post #7

I clicked hoping this would be about how old graphing calculators are generally better math companions than a phone. The best way to do math on my phone I know of is the HP Prime emulator.

PCalc -- because it runs on every Apple platform since the Mac Classic:

https://pcalc.com/mac/thirty.html

My other favorite calculator is free42, or its larger display version plus42

https://thomasokken.com/plus42/

For a CAS tool on a pocket mobile device, I haven't found anything better than MathStudio (formerly SpaceTime):

https://mathstud.io

You can run that in your web browser, but they maintain a mobile app version. It's like a self-hosted Wolfram Alpha.

Re: My iPhone 16 Pro Max produces garbage output when running MLX LLMs

#122

Post by fsckboy » Mon, Feb 02, 2026, 5:51 AM UTC

christophilus wrote: post #79

CrispinS wrote: post #73

Earlier quoted context omitted.

> What's moon plus sun? Eclipse, obviously.

That’s sun minus moon. Moon plus sun is a wildly more massive, nuclear furnace of a moon that also engulfs the earth.

>Moon plus sun is a wildly more massive, nuclear furnace of a moon that also engulfs the earth.

i just looked up mass of sun vs mass of moon (they differ by 10^30 vs 10^20), and the elemental composition of the sun: the moon would entirely disappear into the insignificant digits of trace elements which are in the range of .01 % of the sun. I could be off by orders of magnitude all over the place and it would still disappear.

Re: My iPhone 16 Pro Max produces garbage output when running MLX LLMs

#123

Post by veqq » Mon, Feb 02, 2026, 5:52 AM UTC

Der_Einzige wrote: post #97

dcrazy wrote: post #95

Earlier quoted context omitted.

This thread reminds me of Scribblenauts, the game where you conjure objects to solve puzzles by describing them. I suspect it was an inspiration for Baba Is You.

Scribblenauts was also an early precursor to modern GenAI/word embeddings. I constantly bring it up in discussions of the history of AI for this reason.

Could you explain? :3

Re: My iPhone 16 Pro Max produces garbage output when running MLX LLMs

#124

Post by SauntSolaire » Mon, Feb 02, 2026, 6:01 AM UTC

christophilus wrote: post #79

CrispinS wrote: post #73

Earlier quoted context omitted.

> What's moon plus sun? Eclipse, obviously.

That’s sun minus moon. Moon plus sun is a wildly more massive, nuclear furnace of a moon that also engulfs the earth.

Reminds me of this AI word combination game recently shared on HN, with almost exactly these mechanics:

https://neal.fun/infinite-craft/

For the record, Sun+Moon is indeed eclipse.

Re: My iPhone 16 Pro Max produces garbage output when running MLX LLMs

#125

Post by addaon » Mon, Feb 02, 2026, 6:14 AM UTC

ekelsen wrote: post #109

addaon wrote: post #100

Earlier quoted context omitted.

> Anything that relies on bit patterns of nans behaving in a certain way (like how they propagate) is in dangerous territory. Why? This is well specified by IEEE 754. Many runtimes (e.g. for Javascript) use NaN boxing. Treating floats as a semi-arbitrary selection of rational numbers plus a handful of special values is /more/ correct than treating them as real numbers, but treating them as actually specified does giv…

Can you show me where in the ieee spec this is guaranteed? My understanding is the exact opposite - that it allows implementations to return any NaN value at all. It need not be any that were inputs. It may be that JavaScript relies on it and that has become more binding than the actual spec, but I don't think the spec actually guarantees this. Edit: actually it turns out nan-boxing does not involve arithmetic, which…

Don't have the spec handy, but specifically binary operations combining two NaN inputs must result in one of the input NaNs. For all of Intel SSE, AMD SSE, PowerPC, and ARM, the left hand operand is returned if both are signaling or both or quiet. x87 does weird things (but when doesn't it?), and ARM does weird things when mixing signaling and quiet NaNs.

Re: My iPhone 16 Pro Max produces garbage output when running MLX LLMs

#126

Post by godelski » Mon, Feb 02, 2026, 6:21 AM UTC

geuis wrote: post #89

CrispinS wrote: post #73

Earlier quoted context omitted.

> What's moon plus sun? Eclipse, obviously.

Not obvious. Astronomers are actively looking for signatures of exomoons around exoplanets. So "sun plus moon" could mean that too.

Well you find the signature by looking for a dip in but sun's luminosity. So minus might be the better relationship here

Re: My iPhone 16 Pro Max produces garbage output when running MLX LLMs

#127

Post by mungoman2 » Mon, Feb 02, 2026, 6:22 AM UTC

Good article. Would have liked to see them create a minimal test case, to conclusively show that the results of math operations are actually incorrect.

Re: My iPhone 16 Pro Max produces garbage output when running MLX LLMs

#128

Post by docfort » Mon, Feb 02, 2026, 6:28 AM UTC

Interesting post, but the last bit of logic pointing to the Neural Engine for MLX doesn’t hold up. MLX supports running on CPU, Apple GPU via Metal, and NVIDIA GPU via CUDA: https://github.com/ml-explore/mlx/tree/main/mlx/backend

Re: My iPhone 16 Pro Max produces garbage output when running MLX LLMs

#129

Post by djmips » Mon, Feb 02, 2026, 6:43 AM UTC

bri3d wrote: post #2

I love to see real debugging instead of conspiracy theories! Did you file a radar? (silently laughing while writing this, but maybe there's someone left at Apple who reads those)

IKR - this is very typical

Re: My iPhone 16 Pro Max produces garbage output when running MLX LLMs

#130

Post by ekelsen » Mon, Feb 02, 2026, 7:10 AM UTC

addaon wrote: post #125

ekelsen wrote: post #109

Earlier quoted context omitted.

Can you show me where in the ieee spec this is guaranteed? My understanding is the exact opposite - that it allows implementations to return any NaN value at all. It need not be any that were inputs. It may be that JavaScript relies on it and that has become more binding than the actual spec, but I don't think the spec actually guarantees this. Edit: actually it turns out nan-boxing does not involve arithmetic, which…

Don't have the spec handy, but specifically binary operations combining two NaN inputs must result in one of the input NaNs. For all of Intel SSE, AMD SSE, PowerPC, and ARM, the left hand operand is returned if both are signaling or both or quiet. x87 does weird things (but when doesn't it?), and ARM does weird things when mixing signaling and quiet NaNs.

I also don't have access to the spec, but the people writing Rust do and they claim this: "IEEE makes almost no guarantees about the sign and payload bits of the NaN"

https://rust-lang.github.io/rfcs/3514-float-semantics.html

See also this section of wikipedia https://en.wikipedia.org/wiki/NaN#Canonical_NaN

"On RISC-V, most floating-point operations only ever generate the canonical NaN, even if a NaN is given as the operand (the payload is not propagated)."

And from the same article:

"IEEE 754-2008 recommends, but does not require, propagation of the NaN payload." (Emphasis mine)

I call bullshit on the statement "specifically binary operations combining two NaN inputs must result in one of the input NaNs." It is definitely not in the spec.