Google's Gemma 4 AI can run on smartphones, no Internet required

[DragonSlayer101]

In a nutshell: Google has released the Gemma 4 open-weight AI model, designed to run locally on smartphones and other consumer devices. Built on Gemini 3, Gemma 4 comes in four versions optimized for different use cases, giving users and developers the flexibility to choose the model that best fits their needs.

The two largest Gemma 4 models – 26B Mixture of Experts and 31B Dense – require an 80GB Nvidia H100 GPU to run unquantized in bfloat16 format. Google claims these models deliver "frontier intelligence on personal computers" for students, researchers, and developers, providing advanced reasoning capabilities for IDEs, coding assistants, and agentic workflows.

The 26B model activates only 3.8 billion of its 26 billion parameters in inference mode, resulting in higher tokens-per-second performance compared with similar models, while significantly reducing latency. In contrast, the 31B model focuses on "maximizing raw quality" and allows developers to fine-tune it for specific use cases.

The variants most relevant to end-users are Effective 2B and Effective 4B. These models can run entirely offline and use minimal memory during inference, with only 2 billion and 4 billion parameters, respectively. Google says that reducing the number of active parameters enables these models to run on mobile and IoT devices, including smartphones, Raspberry Pi, and Jetson Nano.

Google claims that its Gemma 4 models are not only significantly faster than Gemma 3 but also the most capable AI models ever designed to run on local hardware. Independent testing appears to support this claim: the 31B model currently ranks #3 on the Arena AI leaderboard for open models, behind GLM-5 and Kimi 2.5, while the 26B sits at #6.

Gemma 4 has been released under an Apache 2.0 license, allowing developers to integrate it into their apps and services without usage restrictions. By comparison, Gemma 3 is governed by a custom Google license with strict usage policies and numerous limitations, making it less attractive for developers.

It is worth noting that, despite the Apache 2.0 license, Gemma 4 is "open-weight" rather than fully open-source. According to the Open Source Initiative, an AI model can only be considered open-source if the complete dataset used for training, along with scripts, infrastructure code, and detailed methodologies, is released.

Google, however, is only releasing the model parameters, not the full, reproducible training pipeline, which prevents others from recreating the model from scratch. For most developers, this limitation is unlikely to matter, as the Apache 2.0 license still permits all forms of commercial use, modification, redistribution, and deployment, with only attribution required.

Permalink to story:

Google's Gemma 4 AI can run on smartphones, no Internet required

 

[Fuzzy Dunalap]

Yes...

 

[Plutoisaplanet]

Does this mean Google plans to copy Microsoft and turn Android into Windows 11?

 

[unoficialoficial]

The 31B variant of Gemma4, which I have tested, is a very capable model, especially in STEM and programming tasks. Let's hope Google will release a bigger MOE variant, such as a 144B A12B, which will set a new frontier standard for the open mid-tier models, much like the QWEN3.5 series did. However, I still like the colorful Gemma2 character.

 

[daffy duck]

Meanwhile the clowns at Adobe won't allow any generative AI to run locally despite how good our hardware is now, just so they milk the sheep for egregious profits.

 

[Axiarus]

It can, but will it on my phone? That is a big hell no.

 

[dangh]

Having it on phone is bit much as it will suck battery life. But on self hosting system it's excellent.

 

[hamza302]

This is actually a pretty big shift in how AI is evolving. Most people are used to models like ChatGPT or Gemini needing constant internet access, but Gemma 4 flips that idea by running directly on the device. That’s huge for privacy and latency—no cloud dependency means faster responses and less data sharing.


What’s interesting is that the smaller Gemma 4 models are specifically designed for mobile and edge devices, and they can run completely offline with near-zero latency on phones and similar hardware:
https://deepmind.google/models/gemma/gemma-4/


From what I’ve seen, Google is clearly pushing toward “local-first AI.” The fact that these models can handle things like reasoning, coding, and even audio/visual tasks on-device shows how quickly efficiency is improving.


If this trend continues, we might reach a point where everyday apps have built-in AI that doesn’t rely on servers at all. That could completely change how we think about AI—more like a standard phone feature rather than a cloud service.