Melvin Johnson (@melvinjohnsonp) 's Twitter Profile
Melvin Johnson

@melvinjohnsonp

Researcher @ Google Research. Multilingual NLP and MT. Previously, Stanford CS.

ID: 2833321760

calendar_today26-09-2014 23:15:53

207 Tweet

2,2K Followers

315 Following

Melvin Johnson (@melvinjohnsonp) 's Twitter Profile Photo

We’re excited to show Deep Think 2.5 Pro which pushes further on inference time scaling to achieve much better performance on hard Math and Coding tasks. The team worked hard burning the midnight oil to get this one out at IO.

Melvin Johnson (@melvinjohnsonp) 's Twitter Profile Photo

Our latest update to Gemini 2.5 Pro is here. It's SoTA on GPQA Diamond, AIDER and HLE. The team has also worked hard to improve the model on style, persona and creativity. We're excited to see what you build with it. Please let us know any feedback as we're eternally cooking.

Melvin Johnson (@melvinjohnsonp) 's Twitter Profile Photo

You go down a little only to come back stronger. Lots of progress ahead from us. Excited for the relentless model improvements coming.

Melvin Johnson (@melvinjohnsonp) 's Twitter Profile Photo

The Gemini 2.5 Pro and Flash models are stable now. We want to thank you all for the feedback and excitement around these two models and can't wait to see what you'll build with it. And we have a new child in the family: 2.5 Flash-Lite, a low-latency, cost-sensitive alternative.

Quoc Le (@quocleix) 's Twitter Profile Photo

Excited to share that a scaled up version of Gemini DeepThink achieves gold-medal standard at the International Mathematical Olympiad. This result is official, and certified by the IMO organizers. Watch out this space, more to come soon! deepmind.google/discover/blog/…

Melvin Johnson (@melvinjohnsonp) 's Twitter Profile Photo

Congrats Thang Luong for your incredible conviction and leadership for #imo2025! You did an amazing job bringing the team together and building the model in a way that would be maximally useful.

Melvin Johnson (@melvinjohnsonp) 's Twitter Profile Photo

We are excited to release Gemini 2.5 DeepThink to Ultra subscribers today. It’s a faster version of the model that won IMO gold. It’s especially useful for hard tasks that require significant thinking and creativity. It shines on hard benchmarks like HLE and LCBv6. Enjoy!