Google veröffentlicht die multimodale Gemma-4-Familie unter Apache-2.0-Lizenz.

Die 26B-MoE- und 31B-Modelle verarbeiten 256.000 Token Kontext und unterstützen natives Function-Calling. Die kompakten E2B- und E4B-Varianten laufen offline auf Edge-Geräten und integrieren dedizierte Audioverarbeitung.

#Google #Gemma4 #LLM #OpenSource #News
https://www.all-ai.de/news/news26top/gemma-4-google-opensource

Gemma 4 schlägt deutlich größere KI-Modelle

Mit vier neuen Modellvarianten bringt Google komplexe Logik auf Smartphones und Workstations.

All-AI.de

Google for Developers (@googledevs)

Google가 엣지 환경에서 실행되는 AI 에이전트 개발을 제안하며, Gemma 4를 #GoogleAIEdge로 활용해 노트북·모바일·IoT 등 다양한 기기에 배포할 수 있다고 소개했다. Gallery 앱에서 실험하거나 LiteRT-LM으로 배포할 수 있어 온디바이스 AI 개발에 유용하다.

https://x.com/googledevs/status/2039802553878094167

#google #gemma4 #edgeai #litetrlm #aiedge

Google for Developers (@googledevs) on X

Go beyond chatbots. Build your next AI agent on your own device. 🤖 Use #GoogleAIEdge to bring Gemma 4's power to the edge. Experiment in the Gallery app or deploy to any device – laptop, mobile, IoT – via LiteRT-LM. What will you build? Learn how: https://t.co/ThUdBiKw14

X (formerly Twitter)

Google for Developers (@googledevs)

Gemma 4 공식 발표 트윗이다. Gemini 3 수준의 연구와 기술을 기반으로 하며, 로컬 하드웨어에서 효율적으로 실행·파인튜닝 가능한 최신 오픈 모델이라고 소개한다. 개발자 대상 고급 추론 기능도 강조한다.

https://x.com/googledevs/status/2039756078506754191

#gemma4 #google #openmodel #llm #local

Google for Developers (@googledevs) on X

Gemma 4 is here! Our most intelligent open models to date, are built on the same world-class research and tech as Gemini 3, and are sized to run and fine-tune efficiently on local hardware. Check out what @GoogleGemma 4 brings to devs: 💎 Advanced Reasoning: Deep logic tasks,

X (formerly Twitter)
Gemma 4: Noua generație de modele AI open‑source

Gemma 4 oferă modele AI open‑source avansate, multimodale și eficiente, optimizate pentru orice hardware și licențiate Apache 2.0.

TECHNEWSRO

merve (@mervenoyann)

Google DeepMind가 로컬 환경에서 실행 가능한 다중 모달 AI 모델 Gemma 4를 여러 크기로 공개했다. 자유 라이선스가 적용되며, transformers, llama.cpp, transformers.js 등 주요 도구에 바로 통합해 에이전트와 함께 플러그앤플레이로 사용할 수 있다고 소개한다.

https://x.com/mervenoyann/status/2039739097611215344

#googledeepmind #gemma4 #localai #multimodal #opensource

merve (@mervenoyann) on X

future is local 🔥 Google DeepMind just released Gemma 4: local frontier in many sizes, all modalities with free license 🤯 we ship Gemma 4 in transformers, llama.cpp, transformers.js and more for your convenience 🫡 plug-and-play with your agents 🙌🏻 read our blog ⤵️

X (formerly Twitter)
Gemma 4: Byte for byte, the most capable open models

Gemma 4: our most intelligent open models to date, purpose-built for advanced reasoning and agentic workflows.

Google

If you've never tried running a model on your own hardware, this is a great
place to start. You pick a model and run it. That's it.

https://perspectiveintelligence.app/?utm_source=mastodon&utm_medium=social&utm_campaign=gemma4_launch

#Gemma4 #OnDeviceAI #PerspectiveIntelligence #OpenSource #Privacy #LocalAI (2/2)

Perspective Intelligence | Unlimited Private AI for Apple Devices

Free, private, on-device AI for iPhone, iPad, and Mac. Chat, transcribe, describe images, and recognize text. All on-device.

Perspective Intelligence

Google выпустила Gemma 4: мощные открытые модели, которые работают локально! 🚀 Лицензия Apache 2.0 позволяет использовать их в любых проектах. От 31B для топовых задач до E2B для смартфонов — теперь ИИ в вашем кармане стал еще умнее и быстрее. Качаем на Hugging Face!

#AI #Gemma4 #нейрончик

Ah, #Google unveils yet another flavor of #AI alphabet soup with "Gemma 4," where you can create #nano #bananas 🍌, #audio #talk to your imaginary friends, and generate #high-fidelity #music that sounds just like your nephew's band in the garage. 🎵 Apparently, this is the future of responsible AI—if your responsibility includes making cinematic cat videos and text-driven selfies. 📸
https://deepmind.google/models/gemma/gemma-4/ #Gemma4 #HackerNews #ngated
Gemma 4

Gemma 4 is a family of open models, purpose-built for advanced reasoning and agentic workflows.

Google DeepMind
Gemma 4

Gemma 4 is a family of open models, purpose-built for advanced reasoning and agentic workflows.

Google DeepMind