Currently trying out Gemma4:e2b and e4b on a six-year old Lenovo Legion 5 with only 6vram using Ollama. e4b is kinda slow, though its fine, but e2b runs great.

#Gemma #Gemma4

#GoogleQuant is a game changer for #LocalAI. Once we see models incorporating it, one can run larger than 30b on 32 GB of #VRAM on processor platforms like #Intel.
Maybe even a 30b qant4 on 24GB of ram on say an #Intel B60 GB.
Google has this advantage to use it in the #GEMMA4 open source series.

New video: Running Gemma 4 on Raspberry Pi 5 with LM Studio CLI, exposing it over local network, and connecting from Zed Editor.
Here's the full walkthrough 👇
youtu.be/kZhAj8--t8w
#RaspberryPi #RaspberryPi5 #Gemma4 #LocalAI #LLM

RichStokoe.BetterTemplates 1.1.0 Templates that have better design principles than the default Microsoft ones.

Zero to fully working local #AI #agents with tools in 2 commands:

> dotnet new btagent
> dotnet run

Part of "Better .NET Templates": nuget.org/packages/Ric...
Github: github.com/richstokoe/b...
Agent Tools nuget: nuget.org/packages/Ric...
#gemma4 #openai #ollama #localai #lmstudio

Gemma 4 Outperforms Frontier Models on Local Benchmarks Gemma 4 outperforms frontier models on FoodTruck Bench with 31B variant. Open model shows production-ready performance for local AI deployment.

Gemma 4 31B just beat several frontier models on real-world benchmarks. It runs locally, admits when it doesn't know things, and costs 4W on edge hardware. Google just changed the local AI game. #locallm #gemma4

bymachine.news/gemma-4-beats-frontier-m...

Gemma 4 Showdown: Linux vs. Mac 🚀

I tested the e4b gemma 4 variant on a 32GB Linux setup vs. a 16GB Mac. The Mac was 4.5x faster (44s vs 199s) and nailed a complex poem constraint. Find more details about _why_ Linux results were different on www.lotharschulz.info/2026/04/04/g... #Gemma4 #AI

جوجل أطلقت Gemma 4 — والخبر نزل من امبارح…
بس أنا مش بتكلم في أي خبر قبل ما أجرّبه بنفسي! 😅
فحمّلت النموذج على لاب توبي
بياناتي محلية 100% مجاناً وبدون إنترنت!

#Gemma4 #Google #LMStudio #الذكاء_الاصطناعي #OpenSource #AIAgent #بدون_انترنت #LLM #حسام_الدين_حسن #epreneurs #ذكاء_اصطناعي_عربي #AIArabic

Just hooked up #Gemma4 with #AntiGravity. Pretty slow, but it works. Now I can code without internet too ^^ 🚀

#ai

Google Launches Open Model Family Gemma 4 Google Launches Open Model Family Gemma 4

#Google launched #Gemma4, a new family of open-weight #AI models. Designed to deliver major advancements in #reasoning, #CodeGeneration & complex #LogicTasks; built using the same research & tech as Gemini 3 but marks a significant leap forward in #capabilities. aibusiness.com/generative-a...

Gemma 4 Finally Works in llama.cpp After Critical Fixes Gemma 4 now runs efficiently in llama.cpp after critical KV cache and tokenizer fixes. Local inference on consumer hardware finally viable.

Gemma 4 is now actually usable in llama.cpp. KV cache and tokenizer bugs fixed. You can run it on consumer GPUs without melting your VRAM. #LocalAI #Gemma4 #LlamaCpp

https://bymachine.news/gemma-4-llama-cpp-fixes-kv-cache

France-Monde - Gemma 4 libère l’IA locale sur smartphones Gemma 4, open source Apache 2.0, déploie IA multimodale hors ligne sur smartphones et Raspberry Pi efficacement.

#Gemma4 open source Apache2.0 IA locale smartphones #Presse83 #IAVar presse83.fr/2026/04/04/g...

Sensazione stranissima, quasi da plot twist di un anime potatopunk di quelliche ti fanno venire voglia di reinstallare tutto da zero e ricominciare:#ai #AIagentica #AIlocalesmartphone #AIondevice #AndroidStudioAI #GeminiNano4 #Gemma #Gemma4 #google
https://www.corrierenerd.it/google-gemma/

Sensazione stranissima, quasi da plot twist di un anime cyberpunk di quelliche ti fanno venire voglia di reinstallare tutto da zero e ricominciare:#ai #AIagentica #AIlocalesmartphone #AIondevice #AndroidStudioAI #GeminiNano4 #Gemma #Gemma4 #google
https://www.corrierenerd.it/google-gemma/

How is Gemma 4?
Has anyone tried it?
#gemma4 #google #ai

oh boy oh boy oh boy - gemma 4 veröffentlicht
na dann will ich nach dem einkaufen mal gucken ob das teil in meinen vram passt
ich war ja schon von qwen3.5 begeistert aber das teil soll noch mal besser sein
bei den open weights gehts momentan richtig ab
#KI #gemma4 #homelab

#google drops a local, free, 26 billion parameters #AI #LLM

#Gemma4

Google just dropped Gemma 4 and it’s a game changer for the local LLM community! 🚀

I tested the new models on my MacBook and Desktop. From vision tasks to complex coding, does it actually live up to the hype?

Watch here: youtu.be/T6AvsQVSL74

#Gemma4 #GoogleAI #LocalLLM

Again, to prove the only #AmericanAI company that stands a chance against #ChineseAI, #Google is constantly releasing #LLM advancements like #TurboQuant & realizes the market shift to #LocalAI accelerating quickly with #OpenSource smaller models.
#Gemma4 is all that.
youtu.be/Kaq5Ual2ij8?...

Google Gemma 4 実践ガイド — Ollama・HuggingFace で動かすマルチモーダル対応オープンモデル - Qiita はじめに 2026年4月2日、Google は Gemma 4 をリリースしました。Gemini 3 の技術をベースにしたオープンウェイトモデルで、Apache 2.0 ライセンスで公開されています。 Gemma 4 の注目ポイントは次の3つです。 マルチモーダル対応...

ローカルで LLM を動かしたいエンジニア必見。Ollama と HuggingFace を使った「Gemma 4」の実践ガイドが話題です。マルチモーダルかつエージェント対応までシームレスに構築できるのは熱い。

個人開発の幅がまた広がりそう。みんなはローカル環境、何で構築してる？🤔

#AI #LLM #Ollama #Gemma4 #エンジニア

https://qiita.com/imk1t/items/1b7fb18b34279d1d4e68

Welcome Gemma 4: Frontier multimodal intelligence on device We’re on a journey to advance and democratize artificial intelligence through open source and open science.

Welcome Gemma 4: Frontier multimodal intelligence on device | HuggingFace blog

huggingface.co/blog/gemma4

#ai #gemma #aimodels #gemma4 #opensource #oss

📝 Daily report 📈

Here are today's most popular trending hashtags #⃣ on our website 🌐️:

#game, #games, #infosec, #video, #pc, #gemma4, #networking, #environment, #nature, #opensuse

🔥 Stay tuned! 🔥

Web記事の重要情報を自動抽出＆要約！リサーチ作業が劇的に変わる 毎日大量のWeb記事を読んで、必要な情報を手作業で抜き出し、要約する作業にうんざりしていませんか？このテンプレは、Playwrightで指定したWeb記事からテキストを抽出し、最新のオープンAIモデルを使って、特定のデータポイントを自動で抽出し、さらに記事全体を要約します。もう情報収集に時間を溶かす必要はありません。

新規事業のリサーチでWeb記事読みまくって情報抽出するの、マジでしんどい。Playwrightで記事内容をぶっこ抜いて、Gemma 4に要約とデータ抽出させるスクリプト書いてる。動的コンテンツのパースでハマりかけたけど、やっと動いた！これで月5時間の作業が5分に✨ #自動化 #業務効率化 #Playwright #Gemma4

📝 詳細はこちら👇
https://note.com/n/nfc6c02010d5a

AIでコメント欄を完全自動モデレーション！不適切投稿の監視からカテゴリ分類まで ユーザーからのコメントや投稿を一つ一つ目視でチェックする作業にうんざりしていませんか？このテンプレは、最新のオープンAIモデルを活用し、WebサイトやSNSのコメント欄から不適切な内容を自動検出し、さらに内容に応じたカテゴリ分類まで行います。もう手作業で時間を溶かす必要はありません。

コメント欄のモデレーション、毎月数時間かかってたのをGemma 4で自動化するコード書いてる。AIに判断させるのが最初は難しかったけど、プロンプト調整で精度90%超え！これで深夜の監視作業から解放される🔥 #自動化 #業務効率化 #Gemma4 #AI

📝 詳細はこちら👇
https://note.com/n/nb6f1d2cbea15

Google released Gemma 4, Apache 2.0 multimodal open models (2B–31B) built for reasoning and agents, optimized for edge and mobile use, long context, and broad developer framework support.

#Gemma4

Ollama output of Gemma4 e4b model with web search tool.

Man laikam patiks #Gemma4 mazākais #AI modelis Gemma4 e2b. Sevišķi ar rīku iespējām, piem ., meklēšanu.
Nevaru sagaidīt, kad beidzot būs laiks aģentu veidošani.

Gemma 4: The new standard for local agentic intelligence on Android News and insights on the Android platform, developer tools, and events.

Gemma 4: The new standard for local agentic intelligence on Android

buff.ly/jXvwofs

#android #ai #gemma #appdev #androidsdk #gemma4 #aimodels

Google's new Gemma 4 lets startups run 31B AI models entirely on local hardware Companies are bleeding cash on cloud API fees. Data breaches are making enterprises paranoid. The global demand for "digital sovereignty" is exploding. So

Google officially launched Gemma 4 on Thursday. The new 31B dense AI model runs entirely on local hardware without cloud servers. #Gemma4 #GoogleAI #TechNews #ArtificialIntelligence

blazetrends.com/googles-new-...

Introducing #Google's #Gemma4: A versatile open AI model designed for devices from smartphones to workstations. Experience advanced reasoning, autonomous workflows, and more. #AI #MachineLearning Link: thedailytechfeed.com/google-launc...

Google Gemma 4 Just Changed the Open-Source AI Game ? Google Gemma 4 is here — open-source, Apache 2.0, and outperforming models 20x its size. Full review, benchmarks, and Gemma 3 vs 4 comparison.

Gemma 4 just raised the bar for open weight AI: 256K context, video understanding, agentic workflows and it still fits on a single GPU.
Full deep-dive on what changed from Gemma 3: techglimmer.io/google-gemma...
#Gemma4 #AI #OpenSource #GoogleDeepMind

ローカルLLM: Gemma 4 をMac Studioで動かしてみる

【Gemma 4をローカルで試す】
Googleのオープンソースモデル「Gemma 4」をOllamaで試したけど、面白い結果に。

31BはMac Studioでも実用外だったけど、モバイル向け「e4b」なら爆速で実用レベル！これ、クラウド経由せずにローカル完結する時代がすぐそこまで来てるな。

みんなはローカルLLM何で動かしてる？🤔

#AI #LocalLLM #Gemma4 #Ollama #エンジニア

https://zenn.dev/hidenori3/articles/611f263e02ced8