Gemma 3 is a collection of lightweight, state-of-the-art open models built from the same research and technology that powers our Gemini 2.0 models. → https://goo.gle/3Dxcg84 Our high-performing foundation models leverage the power of NVIDIA’s GPUs, are available in a range of sizes (1B, 4B, 12B, 27B), and offer capabilities such as support for 140+ languages, multimodal understanding, faster on-device runtime, and a 128K-token context window. We’re also launching ShieldGemma 2, a powerful 4B image safety checker built on Gemma 3. Developers can customize ShieldGemma 2 to suit their safety needs. Ready to start building? Experiment in Google AI Studio and get access to Gemma 3 on Kaggle, Hugging Face, and Ollama.
关于我们
Our goal is to equip developers with the most advanced models to build new applications, helpful tools to write better and faster code, and make it easy to integrate across platforms and devices.
- 网站
-
https://goo.gle/ai-devs
Google AI for Developers的外部链接
- 所属行业
- 科技、信息和网络
- 规模
- 超过 10,001 人
动态
-
?? Native image generation is now available to developers with Gemini 2.0 Flash Experimental. Powerful image generation and conversational image editing is available via the Gemini API. → https://goo.gle/43ExECP Developers can now use Gemini 2.0 Flash Experimental to: ?? Build AI agents that create images ?? Design apps that generate visuals on-demand ?? Brainstorm ideas and quickly prototype visuals Explore image generation with Gemini 2.0 Flash Experimental via the Gemini API in Google AI Studio. → https://goo.gle/420Mxhz
-
-
?? Introducing the Gemma 3 Academic Program, which allows researchers to apply for Google Cloud credits to fast-track their Gemma 3-based research. Applications are now open. → https://goo.gle/4kKo2g0
-
Check out what you can do with Gemma 3 on mobile and web with Google AI Edge. → https://goo.gle/3FsF2av With the new model fully on-device, take advantage of Edge AI benefits like: ? Offline availability ? Cost-efficiency ? Lower latency Get the demo app, download the model from Hugging Face, and start building.
-
-
You asked, we listened. Gemma 3 has some of the most requested features from the community: 16x longer context window (128k), multimodality, 140+ languages, and integrations with your favorite frameworks such as Hugging Face and Ollama → https://goo.gle/3XNAf9M
-
-
ShieldGemma 2 is a turnkey solution for image safety checks based on the new Gemma 3 4B model. → https://goo.gle/3FBMPCI Key features include image safety classifications across 3 categories, easy integration of safety checks into image-based applications, and customizable policy templates. Explore how ShieldGemma 2 enables devs to build safe and responsible AI.
-
Code execution is available in Google AI Studio & the Gemini API, enabling devs to build apps with code-based reasoning. Gemini models can now generate and run Python code in a sandbox, so you can perform calculations, data analysis, and visualizations. → https://goo.gle/3FjARxA
-
-
Check out FastRTC for an easy way to build real-time communication apps in Python. ↓
Building real-time WebRTC and Websocket applications is very difficult to get right in Python. That's why we built FastRTC - the realtime communication library for Python ?? Key features: ??? Automatic Voice Detection and Turn Taking ?? Bring any LLM, audio, or video model ?? Automatic Gradio UI for testing ?? Mount to any FastAPI app for production ?? Call via Phone for Free To get started pip install fastrtc Join and Follow our org: https://lnkd.in/ehvhEnb3 Check out the cookbook https://lnkd.in/e_6fDjAD for conversational apps powered by Google Gemini, SambaNova Systems, ElevenLabs, Groq, Fixie.ai, Kyutai, and OpenAI
-
Explore practical applications of the Gemma model family, including mobile integration, diverse assistant use cases, and specialized fine-tuning techniques. → goo.gle/3QN6ke9
-
Model Explorer's core UI is now a npm package. Visualize large AI models layer-by-layer in your own web apps, with easy integration and powerful insights. Explore these resources ↓ GitHub → https://goo.gle/4in0hc8 npm package → https://goo.gle/3XuFwTC Demo → https://goo.gle/3XzSkIl
-