@yoheinakajima: just ran this on my phone (17 pro)
Summary
RunAnywhere releases an on-device AI benchmark app that lets users measure model performance directly on their own phones, replacing posted numbers with real-time testing.
View Cached Full Text
Cached at: 07/15/26, 01:39 AM
just ran this on my phone (17 pro) https://t.co/uRI98bpYsf
RunAnywhere (YC W26) (@RunAnywhereAI): Every on-device AI benchmark you’ve seen was measured on someone else’s phone.
So we stopped posting numbers and shipped the benchmark instead.
Built into the RunAnywhere app: pick any model, hit Run, and it measures performance on your device.
→ Tokens/sec +
Similar Articles
@rohanpaul_ai: So much possibilities for on-device small models. Here @adrgrondin is running Google’s Gemma 4 E2B on iPhone 17 Pro. ~4…
Google's Gemma 4 E2B is demonstrated running on an iPhone 17 Pro via MLX optimization, achieving ~40 tokens/second with 128K context and offline thinking mode for coding and math.
Run (your largest) local models from your iPhone
A tool or app that enables users to run large local AI models directly from their iPhone, bringing on-device LLM inference to iOS.
@DanKornas: Testing an on-device model shouldn’t mean building a mobile app before you can evaluate it. Google AI Edge Gallery is a…
Google AI Edge Gallery is an experimental beta app that lets developers test and evaluate generative AI models on-device without building a full mobile app. It supports model downloads, custom loading, parameter tuning, hardware benchmarks, and multimodal workflows, and is open-source for Android, iOS, and macOS.
@rohanpaul_ai: Gemma 4 (specifically its edge-optimized E2B and E4B variants) running fully offline on an iPhone via apps like Locally…
Google’s Gemma 4 E2B/E4B quantized variants now run fully offline on iPhone via apps like Locally AI, leveraging the Apple Neural Engine for on-device inference.
I made this android app which runs ai models locally
A developer created an Android app that runs AI models locally, supporting GGUF and LiteRT formats with multiple ways to add models.