Omar Sanseviero
@osanseviero
MTS at @GoogleDeepMind
Building Gemini, Gemma, AI Studio and more. My views
ex-Chief Llama Officer @huggingface 🇵🇪🇲🇽
2.9K Following 72.7K Followers
Congrats Clem, Thom, Julien, Lysandre, and the entire Hugging Face team. Proud of all of you!
The impact of HF in the AI ecosystem has been massive, and I'm so so excited to see them scale that impact with NVIDIA
Congrats @JensenHuang, the goats of ML are joining your team :)
Show more
Today we're releasing TimesFM 3.0 on Hugging Face
- Open foundation model for time series forecasting
- Complex multivariate scenarios with zero-shot
- 330m parameters
Excited about the research and new use cases this will unlock!
Show more
Wild how much generative AI has progressed in the last 3 years
Back then, we were all having fun with Illusion Diffusion and spaghetti celebrities. Where will we be in 3 years from now?
It was great to collaborate with ComfyUI for day 0 support of Omni Flash 1.1
Excited to see what the Comfy community builds with it!
Gemini 3.5 Transcribe is live on AI Gateway with automatic detection for 85+ languages.
Live audio + recordings:
• 𝚐𝚘𝚘𝚐𝚕𝚎/𝚐𝚎𝚖𝚒𝚗𝚒-𝟹.𝟻-𝚝𝚛𝚊𝚗𝚜𝚌𝚛𝚒𝚋𝚎-𝚕𝚒𝚟𝚎
• 𝚐𝚘𝚘𝚐𝚕𝚎/𝚐𝚎𝚖𝚒𝚗𝚒-𝟹.𝟻-𝚝𝚛𝚊𝚗𝚜𝚌𝚛𝚒𝚋𝚎
Show more
How good are agents actually at CAD?
Today we release CADBench, evaluating 10 Models on 105 realistic tasks in Fusion 360
Fable 5 and Gemini 3.7 Flash are #1# with only 24.6% pass rate
Show more
The Gemma community celebrates tonight
Gemma officially achieved 1 billion downloads 🔥
From running underwater to running at space, I'm so excited by how the community is using the models for so many inspiring use cases.
Show more
I asked Gemini 3.7 Flash to write prompts for modern web UI video backgrounds (great challenge to see the creativity of any model) and I really love the resulting aeshetics, rendered with Gemini Omni Flash. Some of the prompts I liked best below - which one is your favourite?
Show more
Gemini 3.7 Flash Developer Guide 🧑💻
Here are a few tips to get the most out of the model:
💭 - Toggle thinking level based on the task
🖼️ - Give screenshots or connect design tools
🤖 - Use Antigravity subagents with model: flash
Read the full details 👇
Show more
Gemini Flash 3.7 was released yesterday and I'm excited to see what folks create with it!
Show us what you've built...and you may get something special🤫👇
Gemini 3.7 Flash is out! ⚡️⚡️⚡️
It's a pretty clear step up from its predecessor in basically all categories we care about.
On APK Arena it sits right between GPT 5.6 Terra and Luna making it a decent contender for mid-core models.
ggs @OfficialLoganK @osanseviero et al!
Show more
SL2T is our breakthrough sign language-to-text model powering new features for Deaf and hard of hearing users on @Android.
Starting with American Sign Language-to-English on Pixel 11, people can sign directly into Gboard and Live Transcribe instead of typing.
Show more
Excited to be in SF next week. Cooking lots of things
Join us in celebrating the upcoming 1 billion downloads of Gemma models🥳
A fun in-person evening with live demos and great people from the open models community
Excited to see you all there!
Show more
Introducing the Gemma Translator project
DIY translation device, powered by Gemma 4 E2B
Using a Raspberry Pi and LiteRT, built with Antigravity, and open source!
We're doing a very special open models event on Aug 20th in SF. DM me if you wanna join!
Maps + Search tool combinations are working now with Gemini.
Here's a small demo of a useful app you can build with the Google Maps tool
