Image created with Flux Pro v1.1 Ultra. Image prompt: CU Boulder brand style — CU Gold & Black, Helvetica Neue, Flatirons, Tuscan-vernacular sandstone + red-tile roofs; Norlin Quad walkway, autumn leaves, 50mm human view, sandstone texture band; integrate the category “Google” via Data Overlay: search-style query UI on a laptop reading “GOOGLE AI”; natural light, clean professional inspiring tone, crisp focus, subtle grain, editorial composition
Exclusive | Perplexity Makes $34.5 Billion Offer for Google’s Chrome Browser – WSJ https://www.wsj.com/tech/perplexity-ai-google-chrome-offer-5ddb7a22
Chatbots Can Go Into a Delusional Spiral. Here’s How It Happens. – The New York Times
This is an important issue but I think this methodology tests how well Claude and Gemini can course correct multiturn ChatGPT conversations rather than how good they are at not getting into the situation in the first place, which is meaningfully different. https://x.com/AmandaAskell/status/1954276447285334151
Matrix-Game 2.0 — The FIRST open-source, real-time, long-sequence interactive world model Last week, DeepMind’s Genie 3 shook the AI world with real-time interactive world models. But… it wasn’t open-sourced. Today, Matrix-Game 2.0 changed the game. 🚀 25FPS. Minutes-long https://x.com/Skywork_ai/status/1955237399912648842
RT @Skywork_ai: Matrix-Game 2.0 — The FIRST open-source, real-time, long-sequence interactive world model Last week, DeepMind’s Genie 3 sh…”” / X https://x.com/slashML/status/1955320183976767673
Damn it worked! Genie 3 world –> inpaint UI –> 4x topaz AI upscale –> train 3d gaussian splat You can step inside a painting of Socrates from 1787. Better than any image-to-3d model I’ve seen. I think Google has stumbled upon the killer app for VR — the literal holodeck. https://x.com/bilawalsidhu/status/1954229425199034753
Lmao. We got open source genie ONE WEEK after Google’s announce. Meanwhile, Odyssey has a launch around the corner too. The future is generated, not rendered.”” / X https://x.com/bilawalsidhu/status/1955342603324453305
RT @altryne: This Genie-3 video is mind boggling, especially this edited out part, the airplane collides with the sphere, bounces off, the…”” / X https://x.com/_rockt/status/1955025996547232170
Today we’re going to start testing a new Google Finance, reimagined with AI at its core. You’ll be able to ask detailed questions about the financial world and get a helpful AI response, alongside links to relevant sites on the web. https://x.com/dozenrose/status/1953839311108902971
Gemini app personalizes responses based on past chats, plus new privacy controls https://blog.google/products/gemini/temporary-chats-privacy-controls/
Gemini 2.5 Pro has a 67% winrate against GPT-5 Thinking https://x.com/scaling01/status/1954546677185970271
Gemma 3 270m 4-bit DWQ is up. Same speed, same memory, much better quality: https://x.com/awnihannun/status/1956089788240728467
Gemma 3 270m 4-bit generates text at over 650 (!) tok/sec on an M4 Max with mlx-lm and uses < 200MB: Not sped up: https://x.com/awnihannun/status/1956053493216895406
Gemma 3 270M running on my Pixel 7a! Absolutely crazy (not sped up) https://x.com/1littlecoder/status/1956065040563331344
Google just dropped a new tiny LLM with outstanding performance — Gemma3 270M. Now available on KerasHub. Try the new presets `gemma3_270m` and `gemma3_instruct_270m`! https://x.com/fchollet/status/1956059444523286870
Google releases Gemma 3 270M, a new model that runs locally on just 0.5 GB RAM.✨ Trained on 6T tokens, it runs fast on phones & handles chat, coding & math. Run at ~50 t/s with our Dynamic GGUF, or fine-tune via Unsloth & export to your phone. Details: https://x.com/UnslothAI/status/1956027720288366883
Introducing Gemma 3 270M: The compact model for hyper-efficient AI – Google Developers Blog https://developers.googleblog.com/en/introducing-gemma-3-270m/
Introducing Gemma 3 270M! 🚀 It sets a new standard for instruction-following in compact models, while being extremely efficient for specialized tasks. https://x.com/googleaidevs/status/1956023961294131488
The new Gemma 3 270M is here https://x.com/ggerganov/status/1956026718013014240
Introducing Gemma 3 270M, a new compact open model engineered for hyper-efficient AI. Built on the Gemma 3 architecture with 170 million embedding parameters and 100 million for transformer blocks. – Sets a new performance for its size on IFEval. – Built for domain and adoption https://x.com/_philschmid/status/1956024995701723484
Introducing Gemma 3 270M: The compact model for hyper-efficient AI – Google Developers Blog
https://developers.googleblog.com/en/introducing-gemma-3-270m/
ollama run gemma3:270m Gemma 3 270M is here! Small model that is extremely efficient to run on-device, and designed for fine-tuning to serve specific agentic use-cases!”” / X https://x.com/ollama/status/1956034607373222042
Tired: painting to video Wired: painting to worlds This is a closest glimpse we’ve seen to a real life holodeck https://x.com/bilawalsidhu/status/1953959597301235943
On the big picture: GPT-5 as a model is pretty much on the same curve as the other top labs. I’d expect the usual leapfrogging between Gemini, Claude, OpenAI, & Grok to continue. Where there are some big gains is that GPT-5 seems well-trained for real world tasks in new ways.”” / X https://x.com/emollick/status/1953565365465964668
🤖 From this week’s issue: A Google Cloud blog post illustrating how to implement short-term and long-term memory for AI agents using the Agent Development Kit (ADK) and Vertex AI Memory Bank. https://x.com/dl_weekly/status/1954308710374760684
Genie 3 can basically do one-shot / single-image 3D reconstruction. Turns a 2D painting into an explorable 3D world, and holy crap the fidelity is nuts. No NeRF, no 3D mesh and blows any image-to-3D tech I’ve seen out of the water. https://x.com/bilawalsidhu/status/1954166512475906217
Genie 3 is great, but IMO and IOI gold alone don’t impress me one bit. It’s just RL- maxxing. I want to see real world results. Show me that these results transfer to any other useful task.”” / X https://x.com/scaling01/status/1955052735918670246
You can now share Gemini Applets with a single flip! 🎉 Publish an instantly shareable by link to the internet! For more control, Advanced sharing still opens the Drive share dialog to set access for individuals or groups 🙂 https://x.com/_philschmid/status/1955301288909885705
AI from the people who pioneered AI https://x.com/OfficialLoganK/status/1954196312691196289
Good news: rate limits have doubled for Deep Think in the @GeminiApp Happy prompting : )”” / X https://x.com/OfficialLoganK/status/1955821580237594847
Great news for @GoogleAI Ultra members on @GeminiApp: ⬆️ We just doubled the 2.5 Deep Think daily limit ⬆️ Any errors on our side don’t count against your limit”” / X https://x.com/joshwoodward/status/1955804081437696046
Caught up with the Gemini Deep Research team in NYC, so cool to hear the story of how they created this category, the crazy product momentum, and what comes next. Stay tuned : ) And yes, we are working on the Deep Research API! https://x.com/OfficialLoganK/status/1955345789850628304
New Google AI Studio landing page just dropped, @ammaar and the team are cooking 🔥 https://x.com/OfficialLoganK/status/1954264163347488887
🚀 Small upgrade to our Deep Research capabilities! 1️⃣ Smarter, more insightful reports 2️⃣ Deeper search for richer findings 3️⃣ More accurate information with less hallucination 4️⃣ Modular tools with parallel execution 5️⃣ Multi-modal input support: upload files, images Try it https://x.com/Alibaba_Qwen/status/1955295298957480298
wtf more than half its parameters are embeddings. https://x.com/code_star/status/1956033343465906379
Call me crazy but I think the newer openai models inc. 5 (and gemini, other maxxing models) are nearly unusable for most of my use cases. Far too verbose, extraordinarily confident in their hallucinations, difficult to steer at times but at the same time fairly sycophantic,”” / X https://x.com/jon_durbin/status/1954263916202316001
How Google and NASA are Testing AI for Medical Care in Space | Google Cloud Blog https://cloud.google.com/blog/topics/public-sector/how-google-and-nasa-are-testing-ai-for-medical-care-in-space
(Group) sport mode just dropped for image gen. Introducing Image Remix in @GroupMe – memes and react pics are now multiplayer. Just tap any photo in your group chat and hit “”remix.”” Live now on iOS! Riff, roast, remix, repeat 🔁 https://x.com/mustafasuleyman/status/1953864100460867819
Imagen 4 is now generally available! The update also introduces a new Imagen 4 Fast model, for rapidly generating images at only $0.02 per image. https://x.com/googleaidevs/status/1956035672197771479
GPT-5-high is pretty good at competitive programming Just a 700 point gap in rating between GPT-5 and Gemini 2.5 Pro https://x.com/scaling01/status/1955053949637021732
Our team spent some time benchmarking the GPT-5 models on one-shot document understanding capabilities. Sharing some WIP results 💡: 1️⃣ GPT-5 mini does a good job. From initial testing it edges out Sonnet and Gemini models. 2️⃣ Surprisingly GPT-5 is middle of the pack (and also https://x.com/jerryjliu0/status/1954293351702036712
@SebastienBubeck Heads up, I’m fairly certain that the o3 run being compared to did not have the google search tool – which is important, since Bulbapedia gives the solution in one tool-use call to some puzzles that can take a much longer time if solved without solution info”” / X https://x.com/kiranvodrahalli/status/1956044490885751273
You can run gpt-oss-20B on Google Colab thanks to @pcuenq @reach_vb 🤯 https://x.com/fdaudens/status/1953420511137931342
