Pytorch is indeed used for many models and is a useful open framework. Most of our open models use either @TensorFlow or JAX, which are Google-initiated open frameworks and are also heavily used.
@jeffdean
-
Google releases 999 open models on Hugging Face
By
–
Check out the 999 open models that Google has released on @huggingface
: https://
huggingface.co/google (Comparative numbers: 387 for Microsoft, 33 for OpenAI, 0 for Anthropic). -
Ancient Greek letter encoding system using five-bit fire signals
By
–
Early long-distance communication system using 5 bits (fire high or low) to encode the 25 Greek letters, one character at a time.
-
Open Buildings Dataset Enabling Healthcare Delivery in Africa
By
–
A nice discussion about how our open buildings dataset is being used to enable healthcare delivery in Africa
-
Gemini 2.5 Flash: High Quality and Lightning Fast Model
By
–
Gemini 2.5 Flash is a high quality model and it's also… quick as a flash. Quick in more ways than one is the best kind of quick.
-
Gemini 2.5 Pro Deep Think: Enhanced Problem-Solving Through Inference Exploration
By
–
More exploration at inference time gives our Gemini 2.5 Pro Deep Think model the capability to solve quite complex problems.
-
Gemma Models Progress: Medical, ASL, and Dolphin Applications
By
–
Some great progress on open versions of Gemma models, whether you care about medical applications, American Sign Language, or dolphin.
-
Gemma Multimodal Models Progress on American Sign Language Translation
By
–
Here's one good sign: there's been some really nice progress on adapting our open Gemma multimodal models to translate American Sign Language to English. https://t.co/FBUvHoKMfO
— Jeff Dean (@JeffDean) 21 mai 2025Here's one good sign: there's been some really nice progress on adapting our open Gemma multimodal models to translate American Sign Language to English.
-
Try Gemini AI Tool – Shared Link Available
By
–
Try it at: https://
g.co/gemini/share/3
e2c4258a61f
… -
Implicit Token Caching for Repeated Input Contexts Now Available
By
–
Caching of tokens when you use the same input context repeatedly now happens implicitly, making things a lot easier to deal with.