Omar Sanseviero

Gemini 2.5 Flash-Lite is now stable and GA!🚀 It is our most cost-efficient, fastest, and cheapest 2.5 model, priced at $0.10/1M input and $0.40/1M output tokens. It has lower latency than 2.0 Flash and shows great performance on a wide range of tasks! developers.googleblog.com/en/gemini-25-f…

Using Adaptive Engine, @SKtelecom tuned open models as small as Gemma 3 4B to exceed frontier performance (GPT-4.1, 3.7 Sonnet, and o4-mini) at multilingual content moderation. Our research 📃 and full results 👇

Official results are in - Gemini achieved gold-medal level in the International Mathematical Olympiad! 🏆 An advanced version was able to solve 5 out of 6 problems. Incredible progress - huge congrats to @lmthang and the team! deepmind.google/discover/blog/…

Build powerful resume agents with our new @crewAIInc and Gemini 2.5 quickstart, using built-in grounding with Google Search to power your Crew with a single API key. In the example: 🔎 Research your public GitHub profile 🔬 Deep dive into your projects 📝 Write a custom CV

Check out this cool MedGemma demo! huggingface.co/spaces/google/…

Fine-tune Gemma3n on videos with audios inside with Colab A100 🔥 Just dropped the notebook where you can learn how to fine-tune Gemma3n on images+audio+text at the same time!

📣 Gemini CLI roadmap It has been so cool seeing everyone building with the Gemini CLI (60k ⭐s!), and sharing feedback (1k open issues 😬) To make things even more transparent the team has made the 🗺️ roadmap public. Take a look and tell us what you think:…

A new walkthrough for a research agent using @llama_index and @Google Gemini, fresh out the oven 🥨 Given a topic: 🌎 Use Gemini 2.5 pro with its server side google search tool 📝 Create an agent that takes notes as it gets results from its websearch 👀 Create other agents that…

Today we are rolling out our first Gemini Embedding model, which ranks #1 on the MTEB leaderboard, as a generally available stable model. It is priced at $0.15 per million tokens and ready for at scale production use!

Next week we're doing an Open Models Meetup in Bangalore and we're looking for speakers! Speakers include team members from Google DeepMind, so you'll hear about Gemma, synthetic data, and architectural evolutions. See you there! Call for Speakers👉 forms.gle/zrWN95Cspmtby4…

Walk the fashion runway with AI in this project from @NSTiwari21 and @margaretmz. Sketch2Runway uses Gemini 2.0 Flash and Veo 3 to enable all levels of fashion designers to transform fashion sketches into runway videos ↓ margaretmz.medium.com/fashion-sketch…

MedSigLIP: create embeddings for medical images and text - 400M text + 400M vision encoder - Useful for classification, semantic image retrieval, and more -Trained with chest X-rays, CT slices, MRI slices, dermatology images, and more. huggingface.co/google/medsigl…

Introducing GenAI Processors ✨ An open source library to build real-time projects easily, with cool features such as stream-based I/O and chaining, modularity, composability, and more GitHub: github.com/google-gemini/… Blog: developers.googleblog.com/en/genai-proce…

Excited to introduce GenAI Processors! An Open-Source Python library from @GoogleDeepMind that allows you to build asynchronous and composable AI Pipelines for Generative AI

New Agent Example! Turn any research question into a data visualization, automatically using Gemini 2.5 Pro and @CamelAIOrg's OWL framework. Exciting collaboration! 🚀 🔍 Performs live web research using search engines and browser use. 🐍 Autonomously writes and executes Python…

Introducing T5Gemma: the next generation of encoder-decoder/T5 models! 🔧Decoder models adapted to be encoder-decoder 🔥32 models with different combinations 🤗Available in Hugging Face and Kaggle developers.googleblog.com/en/t5gemma

Introducing MatFormer Lab for Gemma 3n 🧑‍🔬 Use Mix-n-Match to slice the E4B and create a model with a custom size between 2B and 4B effective parameters Explore the quality-size trade-off and share your models with the community Try it out: goo.gle/gemma3n-matfor…

Gemini API now supports Batch Mode 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 standard API cost. - Receive results within a 24-hour window. - Supports built-in tools like Google…