Research Scientist @googledeepmind, Gemini omni | past: founding researcher @reve and Google brain
Mountain View, CA
- Tweets 224
- Following 468
- Followers 1,080
- Likes 1,640
Come join us and work on omni modality🚀
We are hiring research scientists and engineers🧑🔬 on the Gemini Omni team! Both in the US (Mountain View / SF) and Europe (London / Zürich).
Apply here: google.com/about/careers/app…
17
1,825
Lukasz was leading RL work in brain research back then when transformers were 200m params. It was a privilege seeing his work up close.
Let me take this opportunity to pay tribute to the RLSlow team :) It was a priviledge to lead it through the years, first with @ilyasut then with @merettm and finally on my own - the best team you can ever wish for. It feels like ages ago when we started working on the foundations of reasoning (later berry) with @TrapitBansal, Francis Song, @_smileyball and @sidorszymon.
Many more things happened since then, huge thank you to all teammates through the years: Peter Zhokhov, @unsafePerformIO @vladi_petroff @masonmeyer_ @afrozenator @lukasz_kondr @dmdohan @wang12tao @TheRealRPuri @edwardjhu @eqhylxx @suchirbalaji and Dan Selsam for all the help.
I hope this great technical work continues to be truly and broadly beneficial.
nitter.net/merettm/status/2096630…
1
4
95
15,561
⚡️⚡️⚡️
Same experience, really capable model for day to day coding/research tasks
been using flash 3.8 for a few weeks, given its speed and capabilities it's really hard to choose other models for most daily tasks
you get the same feeling when you had slow internet back in the 90s and suddenly got access to broadband
1
4
687
If you think looping transformers are "new", I refer you to @m__dehghani et al amazing work on universal transformers, published shortly after the transformer paper itself came out. arxiv.org/abs/1807.03819
3
26
230
30,819
If you think SWA's effective receptive field is num_layers*sliding_window_size I have some bad news for you
2
1
48
28,280
Lots of cool research has gone into gemini flash 3.7, I really enjoy using it with @antigravity. Blazing fast and great context handling.
1
16
1,752
Mohammad Saffar retweeted
it’s surprising to me how many people seem to not understand that great models are built with super high quality curated data
finding novel ways to create / get this data is a huge edge
385
211
4,831
713,647
It was a blast working on building Gemini Omni!
We’re dropping Gemini Omni: our first step towards a model that can create anything from anything - starting with video.
It combines Gemini’s intelligence with our generative media systems - representing a leap forward in world understanding, multimodality, and editing 🧵
1
23
10,600
Another banger! A good perk of being at GDM is that we get priority access to Sander 😅
My first blog post in over a year is a deep dive on flow maps🗺️, or how to learn the integral of a diffusion model to enable faster sampling and several other cool tricks.
It's the longest one yet👀 Let me know what you think!
sander.ai/2026/05/06/flow-ma…
1
6
607
Time and time again, the ideas that survive the hype are those that simplify, not overcomplicate.
1
5
377
Reasoning models showed increasing your inference time flops is good. Diffusion LLMs increase inference compute drastically but makes it parallelizable. Diffusion is a promising paradigm for all modalities indeed!
Mercury 2 is live 🚀🚀
The world’s first reasoning diffusion LLM, delivering 5x faster performance than leading speed-optimized LLMs.
Watching the team turn years of research into a real product never gets old, and I’m incredibly proud of what we’ve built.
We’re just getting started on what diffusion can do for language.
1
10
911
