Gemma 4 — Google DeepMind
deepmind.google · 240 words · saved by 1 readers
Our most intelligent open models, built from Gemini 3 research.
Gemma 4 Our most intelligent open models, built from Gemini 3 research and technology to maximize intelligence-per-parameter Slide 1 of 2 Maximum compute and memory efficiency A new level of intelligence for mobile and IoT devices Unprecedented intelligence-per-parameter Frontier intelligence on personal computers Maximum compute and memory efficiency A new level of intelligence for mobile and IoT devices Unprecedented intelligence-per-parameter Frontier intelligence on personal computers Capabilities Performance Safety Download Slide 1 of 5 Agentic workflows Build…
saved by
related reading
- Prime Agent: A self-improving RLM agentprimeintellect.ai
- PostTrainBenchposttrainbench.com
- Pangram 4 Technical Reportpangram-public.s3.us-east-1.amazonaws.com
- google/diffusiongemma-26B-A4B-it · Hugging Facehuggingface.co
- GPT-4openai.com
- gpt-4.pdfcdn.openai.com
- dealignai/Gemma-4-12B-it-qat-JANG_4M-CRACK · Hugging Facehuggingface.co
- Gemma 2: Improving Open Language Models at a Practical Sizearxiv.org
- Together AI | The AI Native Cloudtogether.ai
- AI Model & API Providers Analysis | Artificial Analysisartificialanalysis.ai
- gemini_v1_5_report.pdfstorage.googleapis.com
- VaultGemma: The world's most capable differentially private LLMresearch.google