Vikas Raunak

Research Scientist @GoogleDeepMind. Building Superintelligence.

[Vikas Raunak @vyraun]

imo Gemini 3.5 Flash is like super intelligence on tap for the real world -- super grateful to have helped build this awesome model! Try it!

[Jeff Dean @JeffDean]

1/ Today at #GoogleIO, we’re releasing Gemini 3.5, our latest family of models combining frontier intelligence with action.

We’re starting by releasing 3.5 Flash, which is built to help you execute complex, long-horizon agentic workflows.

Gemini 3.5 Flash is our strongest model for coding and agent https://t.co/m62cBJhIjJ outscores 3.1 Pro on agentic and coding benchmarks like Terminal-Bench and MCP Atlas, while running 4x faster than other frontier models.

Used in Google Antigravity, 3.5 Flash is even further optimized to be up to 12x faster. It’s a powerful engine to deploy sub-agents that collaborate, run high-frequency iterative loops, and solve real-world problems at scale.

Some highlights we’re excited about 🔽

[Liam Fedus @LiamFedus]

Our failures are our moat.

A scientific paper is a clean repackaging of a messy process of failed syntheses, dead ends, hints of success. That mess is the durable asset: what was tried, what worked, what failed, and why.

Compound it into weights you own. Frontier labs live by this principle. If tokens-in-context were enough, pre-training would have died years ago.

[Vikas Raunak @vyraun]

@XYHan_ *compute

[Vikas Raunak @vyraun]

@XYHan_ it has to be this way since we didn't really evolve any special mechanisms for maths/science and basically recycle neuronal organization systems already developed (which computer "intuitions"), Stanislas Dehaene has great empirical work on this

[Vikas Raunak @vyraun]

@lrzneedresearch @emnlpmeeting @aaclmeeting jeez, really bad call by the reviewer. Great research folks would have cared much more about the content. Doesn't matter in the long run, keep doing good work!

[Ilya Sutskever @ilyasut]

🇺🇸🦅250!

[Vikas Raunak @vyraun]

@JohnJumperSci @weihua916 @demishassabis all the best John! amazing run and thanks for all the kind discussions on critical technical problems!

[John Schulman @johnschulman2]

PPO had a second wave in the LLM era for reasons unanticipated by the original paper
- the importance-ratio objective fixes biases from numeric error, async training, and forward pass noise
- the clipping objective affects entropy through a mechanism that we didn't know about at the time of publication (DAPO, https://t.co/sBo9DeFS5Y)

[Vikas Raunak @vyraun]

@NoamShazeer all the best, Noam!

[Vikas Raunak @vyraun]

@gneubig 😂

[Oriol Vinyals @OriolVinyalsML]

1/ Today at Google I/O, we’re launching Gemini 3.5 Flash ⚡️⚡️⚡️!

Our mission was clear: bring frontier-level intelligence with unprecedented speed.

3.5 Flash delivers drastic intelligence (beating 3.1 Pro on almost every benchmark), at Flash speeds. 🧵

[Sundar Pichai @sundarpichai]

Just off stage at #GoogleIO, some highlights from this morning 🧵

Gemini 3.5 Flash is available today for everyone in @antigravity and across our products and APIs.

Compared to 3.1 Pro, 3.5 Flash is better across almost all benchmarks with huge progress in coding. It’s also comparable to the best models but very fast (4x faster tokens/ second than other frontier models). And when looking at the intelligence versus output speed, it’s in a league of its own in the top right quadrant.

[Melvin Johnson @melvinjohnsonp]

Today, we introduced Gemini 3.5 Flash ⚡ Our most capable coding and agentic model — where "fast" and "best" aren't a tradeoff. Try it now across Antigravity, AI Studio, Gemini App, and AI Mode.

[koray kavukcuoglu @koraykv]

Today at Google I/O, we introduced Gemini 3.5 Flash! It has become an integral part of our daily research cycle and works with all the tools we have at Google.
We used a team of agents in Antigravity 2.0 to recreate the original AlphaZero research paper and build a playable version. They coded the reinforcement learning pipeline in JAX/Flax, trained a ResNet model from scratch via self-play on multi-TPU pods, and shipped a full-stack web app so you can play against it, from just 2 prompts. .
Here’s what else makes 3.5 Flash special 🧵

[Google DeepMind @GoogleDeepMind]

Introducing Gemini 3.5: our newest family of models combining frontier intelligence with real-world action.

The first release is 3.5 Flash, our strongest model yet for agents and coding 🧵

[Vikas Raunak @vyraun]

@kaiwei_chang Congrats prof @kaiwei_chang !

[Vikas Raunak @vyraun]

@DimitrisPapail this assumption is likely incorrect -- distillation should amplify these patterns, they are much easier to learn

[koray kavukcuoglu @koraykv]

Incredibly proud to see Alphabet recognized in the 2026 @TIME 100 Most Influential Companies!

Huge congratulations to all the teams across @Google working hard to turn AI breakthroughs into everyday reality for our users! ✨ #TIME100Companies

[Google DeepMind @GoogleDeepMind]

Gemini 3.1 Flash TTS is our most controllable text-to-speech model yet.

With new Audio Tags, you can easily direct vocal style, delivery, and pace through text commands. 🧵

[Vikas Raunak @vyraun]

@AndrewDai @yinfeiy @ElorianAI congrats @AndrewDai & @yinfeiy !