Google Gemma 4 Runs Natively on iPhone with Full Offline AI Inference

#1

Post by takumi123 » Wed, Apr 15, 2026, 5:19 AM UTC

Google Gemma 4 Runs Natively on iPhone with Full Offline AI Inference
gizmoweek.com

Re: Google Gemma 4 Runs Natively on iPhone with Full Offline AI Inference

#2

Post by andsoitis » Wed, Apr 15, 2026, 5:59 AM UTC

is there a comparison of it running on iPhone vs. Android phones?

Re: Google Gemma 4 Runs Natively on iPhone with Full Offline AI Inference

#3

Post by mistic92 » Wed, Apr 15, 2026, 9:52 AM UTC

It runs on Android too, with AI Core or even with llama.cpp

Re: Google Gemma 4 Runs Natively on iPhone with Full Offline AI Inference

#4

Post by bossyTeacher » Wed, Apr 15, 2026, 10:09 AM UTC

Is the output coherent though? I am yet to see a local model working on consumer grade hardware being actually useful.

Re: Google Gemma 4 Runs Natively on iPhone with Full Offline AI Inference

#5

Post by lrvick » Wed, Apr 15, 2026, 10:12 AM UTC

bossyTeacher wrote: post #4

Is the output coherent though? I am yet to see a local model working on consumer grade hardware being actually useful.

I run qwen3.5 122b on a Framework Desktop at 35/ts as a daily driver doing security and OS systems and software engineering.

Never paid an LLM provider and I have no reason to ever start.

Re: Google Gemma 4 Runs Natively on iPhone with Full Offline AI Inference

#6

Post by a_paddy » Wed, Apr 15, 2026, 10:13 AM UTC

bossyTeacher wrote: post #4

Is the output coherent though? I am yet to see a local model working on consumer grade hardware being actually useful.

I can try it for you

Re: Google Gemma 4 Runs Natively on iPhone with Full Offline AI Inference

#7

Post by lrvick » Wed, Apr 15, 2026, 10:13 AM UTC

andsoitis wrote: post #2

is there a comparison of it running on iPhone vs. Android phones?

You can run Android on just about anything so it boils down to Linux GPU benchmarks.

Re: Google Gemma 4 Runs Natively on iPhone with Full Offline AI Inference

#8

Post by jfoster » Wed, Apr 15, 2026, 10:19 AM UTC

bossyTeacher wrote: post #4

Is the output coherent though? I am yet to see a local model working on consumer grade hardware being actually useful.

It can write (some) code that works. Just roughly guessing from my use, but I think of it as being a bit like ChatGPT circa-2024 in terms of capability & speed.

Disappointing if you compare it to anything else from 2026, but fairly impressive for something that can run locally at an OK speed.

Re: Google Gemma 4 Runs Natively on iPhone with Full Offline AI Inference

#9

Post by camillomiller » Wed, Apr 15, 2026, 10:29 AM UTC

[flagged]

Re: Google Gemma 4 Runs Natively on iPhone with Full Offline AI Inference

#10

Post by pabs3 » Wed, Apr 15, 2026, 10:40 AM UTC

> edge AI deployment

Isn't the "edge" meant to be computing near the user, but not on their devices?