DigestAI news desk
Enterprise & Industry updated 1 min read

Older Pixel Smartphones Can Run Advanced AI Models

An enthusiast used a 4-year-old Pixel smartphone, running the Qwen2.5-abliterate:3b model—a 3-billion parameter uncensored AI system—with patience and creativity. The Pixel 7 was modified into a cyberdeck that allowed for on-device AI computation, with a Linux-style terminal display showing commands. Despite limited resources (8GB RAM, Tensor G2 SoC), the smartphone could manage 5 tokens per…

1 source

Key points

  • Pixel 7 modified into a cyberdeck for on-device AI
  • Running Qwen2.5-abliterate:3b model with patience and creativity
  • Limited resources (8GB RAM, Tensor G2 SoC) allow for basic AI tasks
Full story from bing.com · by Omar Sohail · via Search: Google Open source ↗

Older Google Pixel Smartphones Can Become Excellent Hubs For On-Device AI, As One Owner Shows A 4-Year-Old Handset Running An Uncensored 3B Model, With A Catch

bing.com · 12 September 2026

An aging Pixel smartphone will always have multiple use cases, and while Google’s Tensor might be the most underpowered SoC after Huawei’s Kirin series, it’s still capable enough of running smaller AI models if configured properly. For instance, one enthusiast made a cyberdeck housing a Pixel 7 and ran Qwen2.5-abliterate:3b, a 3-billion-parameter uncensored model, which has its safety filters completely removed.

Unfortunately, even running a smaller model on a Pixel 7 requires patience, as the handset only manages 5 tokens per second

Using Claude Code, Redditor “Levfo” generated the 3D-printed CAD files for the cyberdeck case used for the Pixel 7, turning it into a handheld computer capable of running on-device AI. The display shows a Linux-style terminal, displaying the Claude Code command-line tools. The bottom base was likely designed to house a powerbank, allowing the smartphone to run Qwen2.5-abliterate:3b uninterrupted.

Given the limitations of the Tensor G2 running in the Pixel 7 and its 8GB of RAM, running a denser model than one with 3 billion parameters would be an exercise in futility, because even Qwen2.5-abliterate:3b only managed to deliver speeds of 5 tokens per second, which isn’t ideal, but it’s doable if you can muster significant patience. Still, running an AI model without depending on an internet connection is always a perk.

Also, depending on how you view it, running an uncensored model with its safety filters removed can become a positive approach, as long as nothing is harmed in the outcome. While making a cyberdeck version of the Pixel 7 was likely due to a hobby, it shows how older Android smartphones can find purpose once more, thanks to the proliferation of AI.

Of course, a 3B AI model isn’t going to be the most advanced one running on a handheld, but tackling basic tasks shouldn’t be a concern. Instead of keeping such smartphones lying in your drawer or closet gathering dust, would you attempt something similar? Let us know in the comments.

News Source: Reddit

Follow Wccftech on Google to get more of our news coverage in your feeds.

This text was published by bing.com and written by Omar Sohail. It is reproduced here with attribution so you can read it in full; the rights remain with the publisher. Read it at the source ↗

Topics · follow one to build your own front page
GoogleHuaweiQwen2.5-abliterate:3b

The headline, key points and digest above were generated by Digest AI's editorial model from the linked sources. Automated summaries can contain errors: the sources are the record. Spotted a mistake? Tell us.

Comments

via GitHub Discussions

More in Enterprise & Industry

All →

Related stories