@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
@yoheinakajima: local models on your mac, streaming to your phone
RunAnywhere enables users to run AI models on a Mac and stream inference to an iPhone over the local network with one-click setup, avoiding cloud dependency for low latency.
Benchmarking Pocket-Scale Inference
This article benchmarks AI inference performance on the iPhone 17 Pro, evaluating metrics like generation time and model intelligence across various tasks to assess real-world mobile device usage.
@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.
Mobile-use benchmark: which AI model is fastest at using a real Android phone?
A developer built Buddy, an open-source Android app where AI coding agents like Claude Code and Codex control a real phone, and benchmarked which model is fastest at everyday tasks. Sonnet 5.5 and Opus 5.5 topped the results, with Opus notably using a Maps deep link instead of manual tapping.
@googledevs: See how AI models can help you with multi-day engineering workflows with Android Bench 2.0. The updated benchmark evalu…
Android Bench 2.0 is an updated AI evaluation framework that assesses models on long-horizon engineering tasks for Android, such as building apps from scratch and migrating codebases, with continuous scoring.