Video yükleniyor...

Video Yüklenemedi

Ana Sayfaya Dön

Build fully offline mobile assistants with FunctionGemma 270M. This demo parses natural language commands like “Create a calendar event for lunch tomorrow” or “Add to contacts” to execute OS actions directly on-device.

googleaidevs's profile picture

Google AI Developers

116,130 subscribers

86,543 görüntüleme • 7 ay önce •via X (Twitter)

Bilim & Teknoloji
LIVE

Anya Rossi• Live Now

Private livecam show

0 Yorum

Yorum bulunmuyor

Orijinal gönderinin yorumları burada görünecek

Benzer Videolar

Mobile Actions and Tiny Garden are available out-of-the-box in the Google AI Edge Gallery app, now on both Android and iOS. Benchmark the model performance directly in the app. Discover how you can use FunctionGemma (270M) and LiteRT-LM to build private, high-performance local assistants:
0:21

Mobile Actions and Tiny Garden are available out-of-the-box in the Google AI Edge Gallery app, now on both Android and iOS. Benchmark the model performance directly in the app. Discover how you can use FunctionGemma (270M) and LiteRT-LM to build private, high-performance local assistants:

Google for Developers

59,017 görüntüleme • 5 ay önce

Fine-tune FunctionGemma to resolve tool selection ambiguity and build more accurate agentic workflows. The FunctionGemma Tuning Lab lets you train on your own dataset and test natural language to software actions without writing code or managing infrastructure.
0:59

Fine-tune FunctionGemma to resolve tool selection ambiguity and build more accurate agentic workflows. The FunctionGemma Tuning Lab lets you train on your own dataset and test natural language to software actions without writing code or managing infrastructure.

Google AI Developers

20,634 görüntüleme • 6 ay önce

You can now add events to Google Calendar from Gmail with a single click on mobile! After Gemini detects event details in an email, simply tap "Add to Calendar" to save it to Calendar.
0:24

You can now add events to Google Calendar from Gmail with a single click on mobile! After Gemini detects event details in an email, simply tap "Add to Calendar" to save it to Calendar.

Google Calendar

58,999 görüntüleme • 11 ay önce

With the launch of the new #ChatGPT API, I wanted to experiment with language-to-actions, and made a toy app where you use language to build a 3D city! 🌆 This demo is fairly simple but you can imagine eventually just speaking to PowerPoint to auto-build your slides. [1/4]
1:05

With the launch of the new #ChatGPT API, I wanted to experiment with language-to-actions, and made a toy app where you use language to build a 3D city! 🌆 This demo is fairly simple but you can imagine eventually just speaking to PowerPoint to auto-build your slides. [1/4]

Jesse Zhang

700,701 görüntüleme • 3 yıl önce

Google engineer explained how to fine-tune a tiny LLM from 46% to 90% accuracy on your phone in 21 minutes - better than $1500 on-device AI bootcamps. pick Gemma 270M -> generate synthetic task data -> fine-tune with LoRA -> quantize to int4 -> deploy to Pixel and hit 2000 tokens per second. That loop is how a 270M model beats a 70B one on your task, running fully offline in your pocket. Gemma 270M + synthetic data + LoRA + int4 quantization + on-device runtime - that's the stack. Watch and save it, then fine-tune your own tiny agent tonight.
20:45

Google engineer explained how to fine-tune a tiny LLM from 46% to 90% accuracy on your phone in 21 minutes - better than $1500 on-device AI bootcamps. pick Gemma 270M -> generate synthetic task data -> fine-tune with LoRA -> quantize to int4 -> deploy to Pixel and hit 2000 tokens per second. That loop is how a 270M model beats a 70B one on your task, running fully offline in your pocket. Gemma 270M + synthetic data + LoRA + int4 quantization + on-device runtime - that's the stack. Watch and save it, then fine-tune your own tiny agent tonight.

h100envy

704,479 görüntüleme • 1 ay önce

Want to see GitHub Copilot Workspace in action? MV Karan took the stage for a live demo with Copilot Workspace to build a front-end login page, all in natural language.
2:28

Want to see GitHub Copilot Workspace in action? MV Karan took the stage for a live demo with Copilot Workspace to build a front-end login page, all in natural language.

GitHub

107,844 görüntüleme • 2 yıl önce

Starting a group chat is easy! Ready to build your dream team on Element? 1⃣ Open the launchpad and tap “Message” 2⃣ Add contacts with the ‘+’ button or search for friends. (Up to 30) 3⃣ Hit “Create Chat” and you’re in! Perfect for event planning, updates, and staying in the know. Don’t miss out—start growing your community now! 🚀
0:22

Starting a group chat is easy! Ready to build your dream team on Element? 1⃣ Open the launchpad and tap “Message” 2⃣ Add contacts with the ‘+’ button or search for friends. (Up to 30) 3⃣ Hit “Create Chat” and you’re in! Perfect for event planning, updates, and staying in the know. Don’t miss out—start growing your community now! 🚀

Element

22,078 görüntüleme • 2 yıl önce

I made a tiny tool for myself to add events to my gcal, and I've finally decided to ship it (and a silly promo video) cmd + e: AI Event Creation for GCal Add events to gcal with natural language, FAST. Link in the replies
0:57

I made a tiny tool for myself to add events to my gcal, and I've finally decided to ship it (and a silly promo video) cmd + e: AI Event Creation for GCal Add events to gcal with natural language, FAST. Link in the replies

rawn

14,256 görüntüleme • 1 yıl önce

Want to add offline voice AI to your ESP32? Start with ESP-Skainet. It is an open-source speech AI framework for building always-on voice assistants that run entirely on ESP32 devices. It combines WakeNet for wake-word detection, MultiNet for offline speech commands, and an Audio Front End with echo cancellation, noise suppression, and voice activity detection. ESP32-S3 is the recommended platform, with the latest models also targeting ESP32-P4.
0:20

Want to add offline voice AI to your ESP32? Start with ESP-Skainet. It is an open-source speech AI framework for building always-on voice assistants that run entirely on ESP32 devices. It combines WakeNet for wake-word detection, MultiNet for offline speech commands, and an Audio Front End with echo cancellation, noise suppression, and voice activity detection. ESP32-S3 is the recommended platform, with the latest models also targeting ESP32-P4.

Praveen Kumar Verma

15,008 görüntüleme • 28 gün önce

What's the Internet going to look like tomorrow? Chat with AI to create fully functional apps on demand. From personal sites to e-commerce platforms, corporate portals to government services – AI will build it all. The key? Giving AI the right platform to build with.
0:45

What's the Internet going to look like tomorrow? Chat with AI to create fully functional apps on demand. From personal sites to e-commerce platforms, corporate portals to government services – AI will build it all. The key? Giving AI the right platform to build with.

DFINITY Foundation

13,329 görüntüleme • 1 yıl önce

Tomorrow at 12 PM EST holders will be able to activate their Hv-MTLs or claim their AMPs. The website is fully mobile friendly there’s no need for a desktop to participate in this activation. All actions can be completed using a delegate wallet directly from your phone. Below is a walkthrough video that guides you through the entire process.
1:31

Tomorrow at 12 PM EST holders will be able to activate their Hv-MTLs or claim their AMPs. The website is fully mobile friendly there’s no need for a desktop to participate in this activation. All actions can be completed using a delegate wallet directly from your phone. Below is a walkthrough video that guides you through the entire process.

Adam Weitsman

146,680 görüntüleme • 8 ay önce

Claude Code is turning vibe coding into a video game. In a few days will be the easiest way to use Claude Code to build professional mobile apps, and web apps. Use Claude Code in a safe sandbox. Create any interface or application. Create skills and commands. Add integrations and API's. And a lot more. For early access reply below. I'll DM you.
1:17

Claude Code is turning vibe coding into a video game. In a few days will be the easiest way to use Claude Code to build professional mobile apps, and web apps. Use Claude Code in a safe sandbox. Create any interface or application. Create skills and commands. Add integrations and API's. And a lot more. For early access reply below. I'll DM you.

Riley Brown

96,242 görüntüleme • 7 ay önce

Imagine if language models could tap into the app ecosystem of your iPhone. Would the need for plugins and assistants become obsolete if we simply allowed a model to orchestrate our existing (and many years robust) user interfaces? This demonstrates the extent to which GPT-4V excels as a Generalist Mobile AI Agent – without any fine-tuning or grounding, and merely by integrating with a text model that has JSON mode enabled. I suggest watching this demo for a (maybe) wow factor and the results on iOS 17 using NavAIGuide, a mobile and web navigational agent framework for LLMs:
2:03

Imagine if language models could tap into the app ecosystem of your iPhone. Would the need for plugins and assistants become obsolete if we simply allowed a model to orchestrate our existing (and many years robust) user interfaces? This demonstrates the extent to which GPT-4V excels as a Generalist Mobile AI Agent – without any fine-tuning or grounding, and merely by integrating with a text model that has JSON mode enabled. I suggest watching this demo for a (maybe) wow factor and the results on iOS 17 using NavAIGuide, a mobile and web navigational agent framework for LLMs:

Francesco

30,834 görüntüleme • 2 yıl önce

Build a jet lag strategy for your next trip with Gemini 3.5 Flash. Gemini can grab your flight itinerary from Gmail, build a custom sleep schedule based on your time zone, and add it right to your calendar for helpful reminders.
0:28

Build a jet lag strategy for your next trip with Gemini 3.5 Flash. Gemini can grab your flight itinerary from Gmail, build a custom sleep schedule based on your time zone, and add it right to your calendar for helpful reminders.

Google Gemini

35,758,342 görüntüleme • 2 ay önce

Made my first AI executive assistant with Claude code. Claude It can read, send and triage my emails View and create calendar events Create and share Google docs Give me a morning briefing and I can add as many bash scripts for different automations. And yes I can use claude app to chat, it just read some reports on email from linked to google sheets and sent a summary to my email for my next meeting. Now I just book meetings like “find a suitable time on my calendar for this event and send them invites”
0:29

Made my first AI executive assistant with Claude code. Claude It can read, send and triage my emails View and create calendar events Create and share Google docs Give me a morning briefing and I can add as many bash scripts for different automations. And yes I can use claude app to chat, it just read some reports on email from linked to google sheets and sent a summary to my email for my next meeting. Now I just book meetings like “find a suitable time on my calendar for this event and send them invites”

Nelson Amenya

21,692 görüntüleme • 3 ay önce

Build your first AI agent (in 10 minutes): 1) Go to Lindy AI 2) Choose your AI model 4) Select a trigger (calendar event, chat, email, etc.) 5) Add necessary integration steps (Google Calendar, email, etc.) 6) Add AI actions (summarize, research, write emails) 7) Test it out Lindy agents compound over time. What starts as a simple scheduler can become a full-on chief of staff. Get your hands dirty and ship something.
2:48

Build your first AI agent (in 10 minutes): 1) Go to Lindy AI 2) Choose your AI model 4) Select a trigger (calendar event, chat, email, etc.) 5) Add necessary integration steps (Google Calendar, email, etc.) 6) Add AI actions (summarize, research, write emails) 7) Test it out Lindy agents compound over time. What starts as a simple scheduler can become a full-on chief of staff. Get your hands dirty and ship something.

The Startup Ideas Podcast (SIP) 🧃

52,927 görüntüleme • 1 yıl önce

You can now generate videos on This is what I've been using to create my background videos. It's using Runway with a midjourney image. Add prompts like zoom, effects, environments to make the image come to life. Combined with scripting and voices, it's incredibly easy to create shorts or just add b-rolls.
0:43

You can now generate videos on This is what I've been using to create my background videos. It's using Runway with a midjourney image. Add prompts like zoom, effects, environments to make the image come to life. Combined with scripting and voices, it's incredibly easy to create shorts or just add b-rolls.

Meng To

50,513 görüntüleme • 1 yıl önce

The new ability to connect JadeLink, or an SD card reader, to Jade Plus unlocks a powerful new way to sign transactions when using a mobile device, in a fully air-gapped manner. Save the PSBT, sign on Jade, and broadcast with your companion app. Demoed here with nunchuk_io
0:46

The new ability to connect JadeLink, or an SD card reader, to Jade Plus unlocks a powerful new way to sign transactions when using a mobile device, in a fully air-gapped manner. Save the PSBT, sign on Jade, and broadcast with your companion app. Demoed here with nunchuk_io

Blockstream Jade

19,004 görüntüleme • 1 yıl önce