Google AI Edge Gallery
Hacker ApprovedRanked 9 of 27 local LLM runners by recommenders
Free Google app that runs small AI models directly on phones and Macs, working offline and easy to set up.
Free Google app that runs small AI models directly on phones and Macs, working offline and easy to set up.
30 people gave an opinion
- recommend it
- 24
- criticize it
- 6
- approval share
- 80%
Recommended in 9 of 13 months
- November 2025: recommended
- December 2025: recommended
- January 2026: not recommended
- February 2026: not recommended
- March 2026: recommended
- April 2026: recommended
- May 2026: recommended
- June 2026: recommended
- July 2026: recommended
- August 2026: recommended
- September 2026: not recommended
- October 2026: recommended
- November 2026: not recommended (month in progress)
Why it carries the badge
- 24 people recommend itNeeds 10 or more, met.
- 4 recommenders for every criticNeeds 2 to 1 or better, met.
- Recommended in 9 of 13 monthsNeeds 4 or more, met.
The same three rules apply to every product, and the badge can’t be bought. How the badge works
What people say about it
The reasons people give, with the number of comments that give each one. Open a reason to read what they said.
Why people recommend it
Each mark is one comment
Runs AI models on phone offline
“Check out this app they built which loads in the Gemma 270m models and runs them on your phone.”
Comment from 18 Dec 2025 in “FunctionGemma 270M Model” “download the Google Edge Gallery app and one of the chat models, then go into airplane mode”
Comment from 15 Apr 2026 in “CPUs Aren't Dead. Gemma2B Out Scored GPT-3.5 Turbo on Test That Made It Famous” “Your phone can probably run one of these”
Comment from 8 Jun 2026 in “Apple reveals new AI architecture built around Google Gemini models” - Comment from 7 Jul 2026 in “Small AI Models Gain Traction In places with unreliable networks”
Fast on recent phones and Macs
“runs blazingly fast on an M4 iPad or current gen high-end Android device in Google's AI Edge Gallery”
Comment from 19 Apr 2026 in “Android 15's hidden Linux Terminal is a real Debian VM – and it runs Claude Code” “The S25 (edge) runs this very well. 29 tok/s for E2B.”
Comment from 6 Apr 2026 in “Gemma 4 on iPhone” “They run great on android in the Edge Gallery app.”
Comment from 9 Jun 2026 in “Apple bets cheaper AI will woo small developers” - Comment from 3 Jun 2026 in “Gemma 4 12B: A unified, encoder-free multimodal model”
Easy setup, turn-key use
“Anyone with a 16GB Mac — that is quite a lot of journalists, surely — can download that, install a model into it, and play.”
Comment from 3 Jun 2026 in “Gemma 4 12B: A unified, encoder-free multimodal model” “download the Google Edge Gallery app and one of the chat models, then go into airplane mode”
Comment from 15 Apr 2026 in “CPUs Aren't Dead. Gemma2B Out Scored GPT-3.5 Turbo on Test That Made It Famous” “Google Edge Gallery is turn key for people”
Comment from 22 Jun 2026 in “Apertus – Open Foundation Model for Sovereign AI” Useful for emails, scripts, code
“It works perfectly fine for use cases like summarizing emails and such”
Comment from 2 Mar 2026 in “Apple AI servers unused in warehouses due to low Apple Intelligence usage” “very good results for random python scripts”
Comment from 15 Apr 2026 in “CPUs Aren't Dead. Gemma2B Out Scored GPT-3.5 Turbo on Test That Made It Famous” “gave me working code on first try”
Comment from 19 May 2026 in “Gemini 3.5: frontier intelligence with action” - Comment from 25 Mar 2026 in “Local LLM by Ente”
What people warn about
Each mark is one comment
Missing features like web search
“it seems like they hobbled it, not giving access to intents”
Comment from 7 Apr 2026 in “Show HN: Gemma 4 Multimodal Fine-Tuner for Apple Silicon” “it doesn't seem too good at the web search tool”
Comment from 5 Jun 2026 in “Gemma 4 QAT models: Optimizing compression for mobile and laptop efficiency”
Other local LLM runners people recommend
The four most recommended other local LLM runners, with the same counts for each, so you can compare like for like. The tick on each bar marks the 2 to 1 line the badge needs.
| Product | Recommend | Criticize | Approval | Months | Owned | Badge |
|---|---|---|---|---|---|---|
| llama.cpp | 471 | 109 | 81% | 13 | 1 yr4 owners | Approved |
| Ollama | 350 | 247 | 59% | 12 | 1.5 yrs8 owners | — |
| LM Studio | 217 | 74 | 75% | 12 | 3 yrs4 owners | Approved |
| DS4 by antirez | 109 | 20 | 84% | 7 | 1 mo3 owners | Approved |
| Google AI Edge GalleryThis page | 24 | 6 | 80% | 9 | — | Approved |
What people said
29 of the 35 comments that mention it, newest first. Each links to the original comment.
“Blows their Screenshot app out of the water for what I wanted it for.”
“I use Google's own Gallery app instead of Photos.”
“all turned out to be true and good advice”
“Put that on a spare phone”
“Google Edge Gallery is turn key for people”
“You can do that basic fiddling with the Edge AI Gallery app, which can enable thinking and has a customisable system prompt and some agent support.”
“They run great on android in the Edge Gallery app.”
“Your phone can probably run one of these”
“it doesn't seem too good at the web search tool”
“Gemma 4 12B is explicitly listed as unsupported due to not enough RAM even on a 16GB machine”
“on macOS it's very speedy”
“Anyone with a 16GB Mac — that is quite a lot of journalists, surely — can download that, install a model into it, and play.”
“gave me working code on first try”
“Works great in the latest version of Google AI Edge Gallery”
“which is a bloody terrible name for an app by the way”
“runs blazingly fast on an M4 iPad or current gen high-end Android device in Google's AI Edge Gallery”
“very good results for random python scripts”
“download the Google Edge Gallery app and one of the chat models, then go into airplane mode”
“I do remember it not having NPU support (confirmed by other people) back at the Gemma 3 launch a while ago.”
“they just didn't give a f in general. A truly baffling decision.”
“Google Gallery and most third party galleries I've tried lack simple things like adding text on a photo”
“possesses no ransomware capability from what I've last heard of it”
“it seems like they hobbled it, not giving access to intents”
“they're using firebase-analytics and at the very least seem to be sending URLs and infos such as the model you download”
“The S25 (edge) runs this very well. 29 tok/s for E2B.”
“worked really well for me”
“It works perfectly fine for use cases like summarizing emails and such”
“Check out this app they built which loads in the Gemma 270m models and runs them on your phone.”
“I really like the "Google AI Edge Gallery" android app”
Counts come from 1,795,640 Hacker News comments, November 2025 to November 2026. How we count. Report a mistake.