Image created with Flux Pro v1.1 Ultra. Image prompt: Research bullpen with color-block sticky notes forming a knowledge graph; a large title reading “Google” across the top in primary-color sans, clean overlay; query whiteboards outline a Chicago feature package; bright, modern lighting
Today, we’re bringing agentic capabilities to AI Mode in Search for Google AI Ultra subscribers. But… what is actually different? Let’s say you want to make a dinner reservation. Traditionally, that would require multiple searches, concurrent tabs, and a lot of manual https://x.com/GoogleAI/status/1958561117833228705
AI Mode in Google Search adds personalization, agentic features https://blog.google/products/search/ai-mode-agentic-personalized/
Control an avatar in real time (in this case a dog on a beach!) https://x.com/demishassabis/status/1958696900489523633
Genie 3 has advanced spatial memory, when you make changes in the world, they persist in the simulation even when out of view! https://x.com/demishassabis/status/1958696898488840414
Simulations are the future, & one of the main tools we’ll ultimately use to understand and predict things about the universe. This is why I’m so excited about Genie 3, our latest interactive world simulator – here are some insanely cool things you might have missed about it 🧵: https://x.com/demishassabis/status/1958696882105995312
You can prompt Genie 3 with text, photos, or videos – like this cool game example created using Imagen 4 -> Veo 3 -> Genie 3 https://x.com/demishassabis/status/1958696891639595148
Here’s a demo app that shows how you can connect your app to Google Calendar and fetch upcoming events: https://x.com/OpenAIDevs/status/1958660216624751097
GSA, Google Announce Transformative ‘Gemini for Government’ OneGov Agreement | GSA https://www.gsa.gov/about-us/newsroom/news-releases/gsa-google-announce-gemini-onegov-agreement-08212025
Building on our @GoogleWorkspace offer for federal employees, we’re proud to partner with the @USGSA to launch Gemini for Government. More than a model, it’s our complete AI platform with our latest AI tools, including NotebookLM and Veo, powered by our latest models and our”” / X https://x.com/sundarpichai/status/1958538684208476611
We’re bringing more advanced agentic and personalized capabilities to AI Mode — here’s what’s new: 🍝 AI Mode can help you find & book restaurant reservations (coming soon as well for event tickets & local appointments) ☕ Results can now be tailored to your personal https://x.com/Google/status/1958530316072534323
Today we are making URL Context, my favorite Gemini API tool, ready for scaled production use 🔗 The model can now visit webpages, PDF’s, images, and more when you provide the direct URL, and you simply pay for the tokens it processes, no additional tool cost! https://x.com/OfficialLoganK/status/1957551260053225548
Two big updates to the Responses API today. 🖇️ Connectors — Pull context from Gmail, Google Calendar, Dropbox, and more in a single API call. 💬 Conversations — Persist chat threads for your users, without running your own database. More below:”” / X https://x.com/OpenAIDevs/status/1958660207745409120
We can now train AI inside the mind of another AI. 🤯 🌍 Our world model, Genie 3, imagines and generates new worlds on the fly. 🤖 Our embodied agent, Sima, is dropped in and learns to navigate them autonomously. The entire loop—from the environment to the action—is generated https://x.com/bonniesjli/status/1958948293523767561
3d digital twin of a new light rail project in australia bim/cad data + google maps 3d tiles –> interactive 3d engine imagine if every major infrastructure project in the world did this https://x.com/bilawalsidhu/status/1957529179794133147
Gemma 3 270M (8-bit) is also fast on iPhone 16 Pro 🏎️ ~140 tk/s for the A18 Pro chip with MLX, not that far from the M3 chip While Gemma 3 270M is not intended for chat, it’s perfect to be used with Apple Shortcuts on tasks like summarization for example https://x.com/adrgrondin/status/1957171759876059371
Got a Mac with an M-chip? You can now train Gemma3 270m locally as a multilingual embedding or reranker model using our mlx-retrieval project. It lets you train Gemma3 270m locally at 4000 tokens/s on M3 Ultra – that’s actually usable speed. We’ve implemented some standard https://x.com/JinaAI_/status/1958547803489415195
the most sycophantic models are Gemini 2.5 Flash & Pro this explains the lmarena ranking https://x.com/scaling01/status/1956353414687822183
yep, Gemini 2.5 Pro is a glazer “”brilliant, very practical, perfectly, logic is flawless, excellent experimental mathematics”” https://x.com/scaling01/status/1956371713949655328
Well, this happened sooner than I expected… Tencent has dropped their version of Genie 3. https://x.com/bilawalsidhu/status/1955968609940873624
Wow! Chinese lab Tencent Hunyuan has released an open source alternative to Genie 3 🔥 You can generate realistic videos that you can control in real time. – Long-term consistency – No need for expensive rendering – Trained on 1M+ gameplay recordings Already available ↓ https://x.com/itsPaulAi/status/1957182570309013714
Congrats @jparkerholder, @shlomifruchter, & the Genie & Veo teams! If you are interested to know more, the latest episode of the @GoogleDeepMind Podcast with the brilliant @FryRsquared has just dropped, and is all about Genie 3 & its incredible potential: https://x.com/demishassabis/status/1958696904146976927
HTC launched Vive Eagle AI smartglasses in Taiwan, taking on Meta AI glasses The glasses use Google and OpenAI’s models for assistance, and offer similar features such as live photo-based translation Price starts at $520 https://x.com/adcock_brett/status/1957111220474892360
As AI becomes more integrated into our lives, understanding its environmental footprint is essential. ⚡️ That’s why we’re sharing our comprehensive methodology for measuring the energy, emissions, and water impact of Gemini prompts. ↓ https://x.com/GoogleDeepMind/status/1958855573790765273
Measuring the environmental impact of AI inference | Google Cloud Blog https://cloud.google.com/blog/products/infrastructure/measuring-the-environmental-impact-of-ai-inference
Did you know the Gemini API has Batch API? With 50% cost savings! Submit large jobs and retrieve your results within 24 hours at a 50% discount. > Process large batches at 50% of the API cost. > Receive results within a 24-hour window. > Supports built-in tools like Google https://x.com/_philschmid/status/1958910444799726014
Google launches Gemini for Home: New, more powerful AI assistant https://blog.google/products/google-nest/gemini-for-home/
hmm .. gemma3:270m spuriously repeats too often.”” / X https://x.com/kchonyc/status/1956374537278214523
New Gemini Drops just dropped Our regular update to what’s new in the Gemini app is here. 1/ Everything we shipped this month 🧵↓ https://x.com/GeminiApp/status/1956388218217300085
The obvious end result is a serious DeepResearch a la OpenAI/Google. I expected this to debut in V4 but it’s a bit late to the party so naturally we get a trial run on V3 base. https://x.com/teortaxesTex/status/1958751300981604656
90% of Games Developers Already Using AI in Workflows, According to New Google Cloud Research – Aug 18, 2025 https://www.googlecloudpresscorner.com/2025-08-18-90-of-Games-Developers-Already-Using-AI-in-Workflows,-According-to-New-Google-Cloud-Research
Gemini 2.5 Pro, now generally available in @code”” / X https://x.com/code/status/1958238346313863263
This also mention that Gemini is a hybrid reasoning model, i don’t think this was confirmed before”” / X https://x.com/eliebakouch/status/1958603730951029157
Gemini / AI studio REALLY needs a MacOS app. It is time.”” / X https://x.com/bilawalsidhu/status/1957509424312119581
Major TOM @GoogleDeepMind’s AlphaEarth Embeddings are now on @huggingface! 🚀 A new 6 TB prototype dataset for the community. Get it here: https://x.com/mikonvergence/status/1958767622176039019
Masquerade: Learning from In-the-wild Human Videos using Data-Editing https://arxiv.org/pdf/2508.09976
As shown in the livestream, Plus and Pro users globally can now connect Gmail and Google Calendar to ChatGPT for more useful, relevant responses in chat.”” / X https://x.com/OpenAI/status/1956502071756325055
We’re setting up a LOAD of TPUs today and warming them up. Ever wanted to try Veo 3 in @GeminiApp?”” / X https://x.com/joshwoodward/status/1958555951344345461
For a median Gemini text prompt: ⚡️Energy: Less than 9 seconds of watching TV 💧 Water: About 5 drops 🏭 Carbon: 0.03 gCO2e Over a recent 12-month period, we slashed the energy use per prompt by 33x and the carbon footprint by 44x. Find out more ↓ https://x.com/GoogleDeepMind/status/1958855876116455894
🍌”” / X https://x.com/OfficialLoganK/status/1957908528925909391
Generate now: https://t.co/iRl3CSWuvF Generated with Imagen 4 Fast “` { “”scene””: “”minimalist white studio””, “”subjects””: [ { “”type””: “”smartwatch””, “”description””: “”silver frame with red strap””, “”position””: “”center””, “”pose””: “”lying flat”” } ], “”style””: “”photorealistic””, “”lighting””: https://x.com/_philschmid/status/1956351658381705420
Here’s what we shipped this week 🚢🚢🚢 —We launched a new Imagen 4 Fast model so developers can quickly generate images at only $0.02 per image and updated Imagen 4 and Imagen 4 Ultra to support 2K images. All are now generally available in the Gemini API for developers and”” / X https://x.com/GoogleAI/status/1956400937054163357
Imagen 4 is now GA in AI Studio and the Gemini API! Choose from three models: Ultra, Standard, and Fast starting at 2ct per image. ⚡️ Experience up to 10x faster generation than our previous model. 🖼️ Create images with up to 2k resolution. ✍️ Render longer text strings with https://x.com/_philschmid/status/1956351654753673252
Now @GooglePhotos can help you make custom AI-powered edits in seconds — just by asking. You can ask for simple corrective edits — like “remove the cars in the background” — or more creative ones, like adding a party hat or sunglasses. And if you don’t know where to start, try https://x.com/Google/status/1958946812817019305
Google Pixel 10: 9 new AI features and updates https://blog.google/products/pixel/google-pixel-10-ai-features-updates/
It is interesting to see how much effort is going into making ancillary features of the AI models go viral. Ever since the (organic) Studio Ghibli moment, one focus for Grok & Gemini has been on video as a gateway. A challenge has been whether people have creative video ideas.”” / X https://x.com/emollick/status/1956312948130947339
Some fun things people may have missed from Gemma 3 270M: 1. Out of 270M params, 170M are embedding params and 100M are transformers blocks. Bert from 2018 was larger 🤯 2. The vocabulary is quite large (262144 tokens). This makes Gemma 3 270M very good model to be hyper”” / X https://x.com/osanseviero/status/1956258657483534803
Couldn’t resist. Here’s a pure PyTorch from-scratch re-implementation of Gemma 3 270M in a Jupyter Notebook (uses about 1.49 GB RAM): https://x.com/rasbt/status/1957073842393792751
New hyper-efficient addition to our amazing Gemma open models: Gemma 3 270M packs a real punch for its tiny size! It’s super compact and power efficient, so you can easily run your own task-specific fine-tuned systems on edge devices. Enjoy building with it!”” / X https://x.com/demishassabis/status/1956502480675578298
We’re sharing a new addition to the Gemma family of open models: Gemma 3 270M. 🛠️ It’s tiny yet mighty AI – making it ideal for task-specific fine-tuning, with powerful instruction following built-in. Here’s how you can build with it → https://x.com/GoogleDeepMind/status/1956393664248271082
Gemma 3 270M (8-bit) at ~200 tk/s on iPad Air M3 with MLX https://x.com/adrgrondin/status/1956428984876704059
Meet Gemini CLI GitHub Actions, now in beta ✨ Your no-cost AI coding teammate to help automate tasks, review code, and ship faster → https://x.com/googledevs/status/1955615322024657397
Meet Pixel 10, Pixel 10 Pro and Pixel 10 Pro XL. Powered by our new Google Tensor G5 chip and our latest Gemini Nano model, they’re our most personalized, proactive and helpful Pixel phones ever. #MadeByGoogle https://x.com/Google/status/1958218360207921374
You can now share your camera in conversations with Gemini Live to chat back-and-forth about what you see and get real-time advice. @GeminiApp can also point things out too, like what glasses are the best shape for your face¹ 😎 #MadeByGoogle https://x.com/madebygoogle/status/1958216279300403670
Google previews personal health coach for Fitbit https://blog.google/products/fitbit/fitbit-ai-personal-health-coach-preview/
AI efficiency is important. Today, Google is sharing a technical paper detailing our comprehensive methodology for measuring the environmental impact of Gemini inference. We estimate that the median Gemini Apps text prompt uses 0.24 watt-hours of energy (equivalent to watching an https://x.com/JeffDean/status/1958525015722434945
Lucky, lucky ✨ If you’ve been waiting to try #Veo3 tomorrow is your lucky day. Stay tuned… https://x.com/GeminiApp/status/1958558340163699008
Google dropped several products and updates, including: —Memory and temporary chat in Gemini —General availability of Imagen 4, with a fast version —A new Gemma 3 variant with 270M parameter https://x.com/adcock_brett/status/1957111015306350998
URL context | Gemini API | Google AI for Developers https://ai.google.dev/gemini-api/docs/url-context
Over the past few months, we brought AI Mode in Search to the US, India, and the UK. Now, we’re rolling out our most powerful AI search experience to 180+ new countries and territories in English. Excited to expand access to even more languages and regions soon, so stay tuned! https://x.com/rmstein/status/1958552694626607616
The pro models (GPT-5 Pro, Gemini 2.5 Deep Think, Grok 4 Heavy) can be impressive in ways that are hard to see. They take a lot of time to answer questions & are built for very hard problems that require expert evaluation. That is a narrow, but, also very valuable, problem space.”” / X https://x.com/emollick/status/1955902962288746657
Slide into the weekend with 3 video generations at no cost with Veo 3 in the @GeminiApp. But act fast — you have from now until Sunday, 8/24 at 10 pm PT. On your marks, get set, wheee!! 🛝 https://x.com/Google/status/1959037076503937379
Use video generation in @GeminiApp to turn words or photos into videos with sound🔊Type what you want to create to bring any idea to life in minutes.¹ And, if you buy #Pixel10 Pro or Pixel 10 Pro Fold, you get a year of Google AI Pro on us:² https://x.com/madebygoogle/status/1958215989352440270
Build an open-source AI video studio with this @nextjs template using Veo 3 and Imagen 4 in the Gemini API. Create text-to-video, image-to-video, and edit videos in the browser for a specific time range. https://x.com/googleaidevs/status/1958599306472206349
Build your own video generation Studio with Veo 3 in Gemini API! Excited to share an open-source @nextjs template for generating videos with Google’s Veo 3 and Imagen 4. Features: – Generate videos from text prompts using the Veo 3 model. – Create videos from images and text https://x.com/_philschmid/status/1957821851331416079
