BrunoSan KI-PraxisAngewandte KI für Entscheider
← Alle LLMs
LLM · Google

Gemma

Eine Modellfamilie als lebendes Objekt: Varianten, ausgewählte Leistungsdaten und alle verbundenen BrunoSan-Quellensignale.

157Erwähnungen
17in 30 Tagen
65Quellen
17Varianten
Im Universe erkunden →History öffnen →

Modelle

Alle bekannten Varianten dieser Familie. Ein direkter Artikel-Link springt exakt zur genannten Variante.

Google · Gemma

Gemma 4

KNOWN

Technische Leistungsdaten sind für diese Variante noch nicht vollständig bestätigt.

—Kontext
—Max. Output
—Input
—Output
—Input / 1M
—Output / 1M
—Tools
—Reasoning
—Structured
Release/erste Sichtung: 2026-04-02116 Erwähnungen →Universe →
Google · Gemma

Gemma

KNOWN

Technische Leistungsdaten sind für diese Variante noch nicht vollständig bestätigt.

—Kontext
—Max. Output
—Input
—Output
—Input / 1M
—Output / 1M
—Tools
—Reasoning
—Structured
Release/erste Sichtung: 2024-02-2118 Erwähnungen →Universe →
Google · Gemma

Gemma 3n

KNOWN

Technische Leistungsdaten sind für diese Variante noch nicht vollständig bestätigt.

—Kontext
—Max. Output
—Input
—Output
—Input / 1M
—Output / 1M
—Tools
—Reasoning
—Structured
Release/erste Sichtung: 2025-06-264 Erwähnungen →Universe →
Google · Gemma

Gemma 3

KNOWN

Technische Leistungsdaten sind für diese Variante noch nicht vollständig bestätigt.

—Kontext
—Max. Output
—Input
—Output
—Input / 1M
—Output / 1M
—Tools
—Reasoning
—Structured
Release/erste Sichtung: 2025-03-1212 Erwähnungen →Universe →
Google · Gemma

Gemma 3 4B

KNOWN

Technische Leistungsdaten sind für diese Variante noch nicht vollständig bestätigt.

—Kontext
—Max. Output
—Input
—Output
—Input / 1M
—Output / 1M
—Tools
—Reasoning
—Structured
Release/erste Sichtung: 2026-07-231 Erwähnungen →Universe →
Google · Gemma

Gemma 3 27B

KNOWN

Technische Leistungsdaten sind für diese Variante noch nicht vollständig bestätigt.

—Kontext
—Max. Output
—Input
—Output
—Input / 1M
—Output / 1M
—Tools
—Reasoning
—Structured
Release/erste Sichtung: 2026-02-025 Erwähnungen →Universe →
Google · Gemma

Gemma 2 9B

KNOWN

Technische Leistungsdaten sind für diese Variante noch nicht vollständig bestätigt.

—Kontext
—Max. Output
—Input
—Output
—Input / 1M
—Output / 1M
—Tools
—Reasoning
—Structured
Release/erste Sichtung: 2026-06-011 Erwähnungen →Universe →
Google · Gemma

Gemma 2

KNOWN

Technische Leistungsdaten sind für diese Variante noch nicht vollständig bestätigt.

—Kontext
—Max. Output
—Input
—Output
—Input / 1M
—Output / 1M
—Tools
—Reasoning
—Structured
Release/erste Sichtung: 2024-06-275 Erwähnungen →Universe →
Google · Gemma

Gemma 3 1B

KNOWN

Technische Leistungsdaten sind für diese Variante noch nicht vollständig bestätigt.

—Kontext
—Max. Output
—Input
—Output
—Input / 1M
—Output / 1M
—Tools
—Reasoning
—Structured
Release/erste Sichtung: 2026-01-152 Erwähnungen →Universe →
Google · Gemma

Gemma 3 270M

KNOWN

Technische Leistungsdaten sind für diese Variante noch nicht vollständig bestätigt.

—Kontext
—Max. Output
—Input
—Output
—Input / 1M
—Output / 1M
—Tools
—Reasoning
—Structured
Release/erste Sichtung: 2026-03-041 Erwähnungen →Universe →
Google · Gemma

Gemma 2 2B

KNOWN

Technische Leistungsdaten sind für diese Variante noch nicht vollständig bestätigt.

—Kontext
—Max. Output
—Input
—Output
—Input / 1M
—Output / 1M
—Tools
—Reasoning
—Structured
Release/erste Sichtung: 2024-07-311 Erwähnungen →Universe →
Google · Gemma

Gemma-Ko-9B

KNOWN

Technische Leistungsdaten sind für diese Variante noch nicht vollständig bestätigt.

—Kontext
—Max. Output
—Input
—Output
—Input / 1M
—Output / 1M
—Tools
—Reasoning
—Structured
Release/erste Sichtung: — Erwähnungen →Universe →
Google · Gemma

Gemma-Ko-7B

KNOWN

Technische Leistungsdaten sind für diese Variante noch nicht vollständig bestätigt.

—Kontext
—Max. Output
—Input
—Output
—Input / 1M
—Output / 1M
—Tools
—Reasoning
—Structured
Release/erste Sichtung: — Erwähnungen →Universe →
Google · Gemma

Gemma 3n E4B

KNOWN

Technische Leistungsdaten sind für diese Variante noch nicht vollständig bestätigt.

—Kontext
—Max. Output
—Input
—Output
—Input / 1M
—Output / 1M
—Tools
—Reasoning
—Structured
Release/erste Sichtung: — Erwähnungen →Universe →
Google · Gemma

Gemma 3n E2B

KNOWN

Technische Leistungsdaten sind für diese Variante noch nicht vollständig bestätigt.

—Kontext
—Max. Output
—Input
—Output
—Input / 1M
—Output / 1M
—Tools
—Reasoning
—Structured
Release/erste Sichtung: — Erwähnungen →Universe →
Google · Gemma

Gemma 3 12B

KNOWN

Technische Leistungsdaten sind für diese Variante noch nicht vollständig bestätigt.

—Kontext
—Max. Output
—Input
—Output
—Input / 1M
—Output / 1M
—Tools
—Reasoning
—Structured
Release/erste Sichtung: — Erwähnungen →Universe →
Google · Gemma

Gemma 2 27B

KNOWN

Technische Leistungsdaten sind für diese Variante noch nicht vollständig bestätigt.

—Kontext
—Max. Output
—Input
—Output
—Input / 1M
—Output / 1M
—Tools
—Reasoning
—Structured
Release/erste Sichtung: — Erwähnungen →Universe →

BrunoSan Chronik

Alle verbundenen Artikel, neueste zuerst.

01. September 2026

Leverage Android skills and Gemma 4 in Android Studio Quail 4android-developers.googleblog.com

31. August 2026

What’s new in AI infrastructure and orchestration this monthcloud.google.com

28. August 2026

My 2 cents on models pricingcommunity.openai.com

26. August 2026

Agentic AI – Ongoing coverage of its impact on the enterprisecomputerworld.com

25. August 2026

Nvidia says its Groq 3 LPX is four times faster than Cerebras, but the math is more complicatedthe-decoder.com

24. August 2026

Nvidia says its Groq 3 LPX racks delivered 3,400 tokens per second in an Artificial Analysis benchmark running Gemma 4 31B with a 100,000-token input sequence (The Register)techmeme.com
What Nvidia's first Groq 3 LPU benchmarks tell us about its $20B gambletheregister.com
What Nvidia's first Groq 3 LPU benchmarks do and don't tell us about its $20B gambletheregister.com

20. August 2026

Google DeepMind says its Gemma family of open models has surpassed 1B downloads and developers have published 100K+ Gemma model variants over the past two years (Google)techmeme.com

19. August 2026

Tutorial: OSS Kubernetes on GCE with NVIDIA B200 GPUs, DRA for GPU, DRANET (OSS) and Gemma 4 (31B)discuss.google.dev
OSS Kubernetes on GCE with NVIDIA B200 GPUs, DRA for GPU, DRANET (OSS) and Gemma 4 (31B)discuss.google.dev

17. August 2026

I ran the tiny Bonsai model on my tiny GPU. Here’s how it performedinfoworld.com

15. August 2026

I Compared 7 Best Small Language Models for 2026aitechtrend.com

14. August 2026

How to effectively serve MTP-based Gemma 4 models for inference performancediscuss.google.dev

09. August 2026

Google's DiffusionGemma proves you don't need to train from scratch to build a text diffusion modelthe-decoder.com

07. August 2026

Leadership Shake-Up at Google DeepMindblog.google

06. August 2026

Inside Android Skills - Built for deprecationandroid-developers.googleblog.com

02. August 2026

Google’s Gemma 4 open AI models use “speculative decoding” to get up to 3x fasterarstechnica.com

29. Juli 2026

Show HN: Open-source engine running Gemma 4 26B in 2 GB RAM on any M-series Macgithub.com

28. Juli 2026

Show HN: Gemma 4 26B A4B running on an iPhone 17 Pro via model pagingnoemaai.com

27. Juli 2026

AT&T unveils telco open AI modelcomputerweekly.com

26. Juli 2026

Show HN: External KV Cache Offloading Cuts Long Horizon Inference Costs by 50%github.com
Show HN: I built a hypervisor and client for inference on consumer computescalattice.com

23. Juli 2026

NASA Puts Google’s Gemma Large Language Model in Orbitspectrum.ieee.orgIM MAGAZIN

21. Juli 2026

Build intelligent Android apps: On-device inferenceandroid-developers.googleblog.com

13. Juli 2026

Show HN: Biscotti – on-device meeting transcription for macOSnews.ycombinator.com
Gemma gemma-4-26 quota can't be adjusteddiscuss.google.dev

12. Juli 2026

Show HN: Godot running Gemma 4 inference in GDScript and Vulkan compute shadersgithub.com

10. Juli 2026

Show HN: Cactus v2 – On-device AI with cloud fallbacknews.ycombinator.com

09. Juli 2026

Evolving LLM fine-tuning hyperparameters with AlphaEvolve on Google Clouddiscuss.google.devIM MAGAZIN
Autopilot Clusters with GKE managed DRANET: GPUs and TPUscloud.google.com

08. Juli 2026

Gemma 4 Technical Report [pdf]arxiv.org

07. Juli 2026

Ukraine to pick AI models operated without provider control, official sayseconomictimes.indiatimes.com
Google Cloud Labs: Accelerate AI with Cloud Runcloud.google.com
Show HN: Rewire Text – system-wide text transforms with BYOK and local-modelssunsetmesasoftware.com

06. Juli 2026

Small AI Models Gain Traction Around the Worldspectrum.ieee.org
Yesterday I wrote a new coding agent (CodeGoLLM)community.openai.com
When Gemma Thinks About Resources – It Fails: A Behavioral Experimentlesswrong.com

04. Juli 2026

Show HN: Gemma 3 inference in pure C++ with Metal accelerationgithub.com
Show HN: Running a vision model on every screenshot on-devicegithub.com

02. Juli 2026

Ask HN: What are you go to LLM models for the followingnews.ycombinator.com
Gemma 4 in the browser with WebGPUhuggingface.co

01. Juli 2026

Scaling LLM Inference: Multi-Node KV Cache Offloading with GKE & Managed Lustrecloud.google.com
Faster Gemma 4 on MLX with multi-token predictionollama.com
Hugging Face and Cerebras bring Gemma 4 to real-time voice AIhuggingface.co

30. Juni 2026

Show HN: Running Gemma-4 26B at 124 tokens/SEC on a CPU, no GPUapeg.dev
Stop Choosing Between Local and Cloud LLMs: A Field Guide to Hybrid Patternstowardsdatascience.com
Aarogya Setu 2.0 to use Google’s Gemma model to digitise health recordsmedianama.com
Gemma 4 on Cerebras–The Fastest Inference Is Now Multimodalcerebras.ai

29. Juni 2026

DeepSeek details DSpark, a speculative decoding framework for its V4 models, saying it speeds up AI inference by up to 85% and was tested on Gemma and Qwen (Ben Jiang/South China Morning Post)techmeme.com

26. Juni 2026

From Local LLM to Tool-Using Agenttowardsdatascience.com
Visual Studio Code mit lokalem LLM nutzen – so geht‘scomputerwoche.de

24. Juni 2026

CUDA Error: Running Gemma4 12B in Cloud Rundiscuss.google.dev

23. Juni 2026

Build Your Own Local AI Coding Agent with Gemma 4 and OpenCodetowardsdatascience.com

18. Juni 2026

Agent Factory Recap: 100X engineering with AI agents in Google Antigravity 2.0cloud.google.com

15. Juni 2026

AWS Weekly Roundup: AWS FinOps Agent in preview, Gemma 4 on Bedrock, Kiro Pro Max, and more (June 15, 2026)aws.amazon.com

10. Juni 2026

Google’s DiffusionGemma is 4x faster than its other Gemma modelsthenewstack.io

09. Juni 2026

Top 3 updates for Android developer productivityandroid-developers.googleblog.com

04. Juni 2026

Google Gemma 4 12B nearly matches 26B benchmarks — and runs on your laptopthenewstack.io
Google’s Gemma 4 12B brings local multimodal AI to laptopsdeveloper-tech.com

02. Juni 2026

Experimenting with TPUs, GKE Managed DRANET, and Multi-cluster Inference Gatewaycloud.google.com

01. Juni 2026

How Trustpilot built a real-time architecture for data enrichment using Gemmacloud.google.com

27. Mai 2026

A Guide to AI Cold Starts on Cloud Runcloud.google.com

26. Mai 2026

Top AI on Android updates for building intelligent experiences from Google I/O ‘26android-developers.googleblog.com

19. Mai 2026

17 Things to know for Android developers at Google I/Oandroid-developers.googleblog.com
Android Studio I/O Edition: What’s new in Android Developer toolsandroid-developers.googleblog.com

15. Mai 2026

Together AI and Pearl Research Labs Team Up to Reduce the Cost of AI Inferencetogether.ai

11. Mai 2026

Google brings multi-token prediction Gemma 4 LLMsbdtechtalks.com

29. April 2026

Gemma 4 architecture support for QVAC-Fabric (Tether's llama.cpp fork)github.com
Show HN: A new benchmark for testing LLMs for deterministic outputsinterfaze.ai

28. April 2026

How to run a local coding agent with Gemma 4 and Pipatloeber.com
Yesterday I wrote a new coding agentcommunity.openai.com

27. April 2026

Using Google's Gemma 4 E4B Local AI Model to Reverse Engineer a Simple Crackmegithub.com
Running Gemma 4 31B on Mac with Ollamasammyrulez.github.io
Ask HN: Will local models on normal hardware ever compete?news.ycombinator.com
[SOLVED] -if current rates are ok, monthly rates have just warnings for future eventual billing enable!discuss.google.dev
AiStudio/Gemini API Rate Limit/RPD blocking/deblocking periods/checksdiscuss.google.dev

26. April 2026

Show HN: I read Replika's privacy policy and then built a competitorapps.apple.com
A weekend with LoRA on Gemma 4 E2B: instrumenting what fine-tuning changesaiexplr.com

23. April 2026

Day 2 at Google Cloud Next: A marathon developer keynotecloud.google.com

22. April 2026

Gemma 4 VLA Demo on Jetson Orin Nano Superhuggingface.co
Introducing Gemini Enterprise Agent Platform, powering the next wave of agentscloud.google.com
What’s new in GKE at Next ‘26cloud.google.com
Google’s Gemma 4 shines on local systems – both big and smallinfoworld.com

18. April 2026

Gemma 4 Tool Calling Explained: Build AI Agents with Function Calling (Step-by-Step Guide)analyticsvidhya.com

14. April 2026

Gemopus: A Gemma fine-tune that prioritizes stability over long chain-of-thoughthuggingface.co
Gemma 4 and the Economics of Selling AIgertlabs.com
Google Releases Gemini Robotics-ER 1.6 Model To Give Robots Eyes That Can Actually Read The Roomofficechai.com
MiniMax M2.7 Goes Open-Weight to Let You Run Agents Locallyanalyticsvidhya.com
Student researcher blocked: Have TRC TPU quota but no credit card for GCP billing verificationdiscuss.google.dev

13. April 2026

Top 10 Gemma 4 Projects That Will Blow Your Mindanalyticsvidhya.com
Strengthening enterprise governance for rising edge AI workloadsartificialintelligence-news.com

08. April 2026

How to Run Gemma 4 on Your Phone Without Internet: A Hands-On Guideanalyticsvidhya.com
Running Gemma 4 Locally with Ollama on Your PCanalyticsvidhya.com

06. April 2026

Google Unveils Gemma 4: Why tech giant’s latest open source code is buzzing; how it is different from ‘Gemini’etnownews.com

03. April 2026

Google announces Gemma 4 open AI models, switches to Apache 2.0 licensearstechnica.com
A Visual Guide to Gemma 4newsletter.maartengrootendorst.com
Google's Gemma 4 is good, but you won't use itriteshkhanna.com
Google battles Chinese open-weights models with Gemma 4theregister.com
Google DeepMind Opens Up AI With Gemma 4 Family of Four Lightweight Modelsciolookmedia.com
Google gives enterprises new controls to manage AI inference costs and reliabilityinfoworld.com
Google’s Gemma 4 Runs Offline on Any Device, & Is Completely Free to Usetimesofai.com
Running Qwen 35B on 16GB Mac Mini via MMAP (17 tok/s) then swapping to Gemma 4thoughts.jock.pl
April 2026 TLDR Setup for Ollama and Gemma 4 26B on a Mac minigist.github.com
Gemma 4 makes local AI agents practicalfirethering.com
Show HN: Local RAG on 25 Years of Teletext Newsgithub.com
Google lanceert Gemma 4 met een breed licentiemodeltechzine.be
Google launches Gemma 4 with a broad licensing modeltechzine.eu

02. April 2026

Google battles Chinese open-weights models with Gemma 4go.theregister.com
Google’s Gemma 4 Model Can Now Be Deployed on NVIDIA’s RTX GPUs, Delivering Optimized Performance for a ‘Personalized’ Agentic AI Environmentwccftech.com
Gemma 4 reports, configs, and code, here are the takeawaystwitter.com
Gemma 4 running locally in the browser with transformers.jshuggingface.co
Google Launches ‘Truly Open Source’ Gemma 4techstrong.ai
Google unveils Gemma 4, expands lightweight open model lineup for developerseconomictimes.indiatimes.com
Google's Gemma 4 is now available with Apache 2.0 licensing for the first timethe-decoder.com
Google releases powerful Gemma 4 modelsdiscuss.google.dev
Gemma 4 Just Releasedtwitter.com
Google Releases Gemma 4 Open Models, Calls Them “Best In World” In Their Categoryofficechai.com
Gemma 4 running on Nvidia and AMD from Day 0 with MAXmodular.com
Bringing AI Closer to the Edge and On-Device with Gemma 4developer.nvidia.com
From RTX to Spark: NVIDIA Accelerates Gemma 4 for Local Agentic AIblogs.nvidia.com
End-to-end guide: fine-tuning and serving Gemma 4 on Vertex AIdiscuss.google.dev
Google releases Gemma 4 open modelsdeepmind.google
Gemma 4 Model Cardai.google.dev
Google launches Gemma 4, its "most intelligent" open model family, purpose-built for advanced reasoning and agentic workflows, under an Apache 2.0 license (The Keyword)techmeme.com
Google announces Gemma 4 open AI models, switches to Apache 2.0 licensearstechnica.com
Gemma 4blog.google
Google's Gemma 4 model goes fully open-source and unlocks powerful local AI - even on phoneszdnet.com
Google’s new Gemma 4 models bring complex reasoning skills to low-power devicessiliconangle.com
Introducing Gemma 4 on Google Cloud: Our most capable open models yetcloud.google.com
Gemma 4: The new standard for local agentic intelligence on Androidandroid-developers.googleblog.com
Announcing Gemma 4 in the AICore Developer Previewandroid-developers.googleblog.com
Android Studio supports Gemma 4: our most capable local model for agentic codingandroid-developers.googleblog.com
Welcome Gemma 4: Frontier multimodal intelligence on devicehuggingface.co

30. März 2026

Show HN: SycoFact 4B: Open model detecting sycophancy and delusion confirmationhuggingface.co

26. März 2026

10 LLMs, die wenig bis keine Grenzen kennencomputerwoche.de

23. März 2026

Running inference on vllm with Dynamic Resource Allocation and Custom Compute Classesdiscuss.google.dev

12. März 2026

Selectively reducing eval awareness and murder in Gemma 3 27B via steeringlesswrong.com

09. März 2026

19 large language models redefining AI safety—and dangerinfoworld.com

05. März 2026

Show HN: Nemilia – multi-agent AI workspace in a single HTML file, no back endgithub.com
Show HN: Nemilia – multi-agent AI workspace in a single HTML file, no back endgithub.com

04. März 2026

Small models, high quality: Inside BMW Group’s experiments evaluating domain-specific language modelscloud.google.com

02. Februar 2026

High-performance inference meets serverless compute with NVIDIA RTX PRO 6000 on Cloud Runcloud.google.com

28. Januar 2026

How Automated Prompt Optimization Unlocks Quality Gains for ML Kit’s GenAI Prompt APIandroid-developers.googleblog.com

15. Januar 2026

Introducing BigQuery managed and SQL-native inference for open modelscloud.google.com

31. Dezember 2025

What Google Cloud announced in AI this month - 2025cloud.google.com

26. Dezember 2025

What’s new with Google Cloud - 2025cloud.google.com

19. Dezember 2025

The Year in Google Cloud — 2025cloud.google.com

17. November 2025

Production-Ready AI with Google Cloud Learning Pathcloud.google.com

26. Juni 2025

Run Google DeepMind’s Gemma 3n on NVIDIA Jetson and RTXdeveloper.nvidia.com
Gemma 3n fully available in the open-source ecosystem!huggingface.co

24. Juni 2025

Run Open-Source AI Models Locally with Docker Model Runnerblog.agentailor.com

12. März 2025

Welcome Gemma 3: Google's all new multimodal, multilingual, long context open LLMhuggingface.co

26. September 2024

Your ultimate guide to the latest in generative AI on Vertex AIcloud.google.com

31. Juli 2024

Google releases Gemma 2 2B, ShieldGemma and Gemma Scopehuggingface.co

27. Juni 2024

Welcome Gemma 2 - Google’s new open LLMhuggingface.co

21. Februar 2024

NVIDIA TensorRT-LLM Revs Up Inference for Google Gemmadeveloper.nvidia.com
OpenRouter liefert — sofern vorhanden — technische Modelldaten. BrunoSan liefert die Nachrichtenbeziehungen, Erwähnungszahlen und Chronik. Ein OpenRouter-Ausfall entfernt keine BrunoSan-Erkenntnis.