Gabriel Rasskin

Google Keras team. CMU. Co-creator of

[François Chollet @fchollet]

Super simple Colab that shows you how to build a chatbot using Gemma 2 9B: https://t.co/govVZnDPPV

Built with Keras 3, runs on any backend -- JAX, PyTorch, TensorFlow. Personally recommend JAX for best performance.

[Gabriel Rasskin @GabrielRasskin]

Proud to have been a part of this, try it out now in Jax, Tensorflow or Pytorch with the Keras implementation - both 9B and 27B have a lot of personality

[Google DeepMind @GoogleDeepMind]

We're excited to unveil Gemma 2. 🛠️

Available in both 9B and 27B parameters, it delivers the best performance for its size - unlocking more possibilities for developers to build and deploy with AI. → https://t.co/RmRLEdFPMV

[Jeff Dean @JeffDean]

Gemma 2 is now available available in 9B and 27B sizes (great for practical deployments) 🎉

These open-source models offer best-in-class performance (often better than significantly larger models), run at incredible speed across different hardware and easily integrate with other AI tools.

Download the weights on @Kaggle and @HuggingFace, or access the models in Google AI Studio

Both sizes also do very well compared to other proprietary and open models on lmsys (although error bars are still largish due to limited votes so far):
See more on the blog:
https://t.co/0WrtElgh8e

Tech report: https://t.co/RCREbHRIse

Congrats to all the people who have been working hard on this new release!

[François Chollet @fchollet]

We're releasing a new high-performing open LLM: Gemma 2! Written in Keras 3, and compatible with TensorFlow, JAX, and PyTorch.

Gemma 2 comes in two sizes, 9B and 27B parameters, with standard and instruction-tuned variants. Find them here:

On Kaggle: https://t.co/CEhLE5Iayd
On Hugging Face: https://t.co/m0vcdG3Puo

[meg.ai 🇨🇦 @MeganRisdal]

Gemma 2!!!!!!! #GemmaVerse #GoogleIOConnect Berlin!

[Gabriel Rasskin @GabrielRasskin]

Dived into how Anthropic made Golden Gate Claude, and more generally how we can make LLM's do the things we want them to https://t.co/QGidsL3iUO

[François Chollet @fchollet]

If you missed it, this I/O session on LLMs with Keras 3 is a great tutorial on LLM training and fine-tuning best practices https://t.co/QwktS21G9E

[Samaneh Saadat @smn_sdt]

If you're at #GoogleIO and want to chat about Keras, stop by the Office Hour tent! @martin_gorner, @GabrielRasskin and I will be there!

[Gabriel Rasskin @GabrielRasskin]

#GoogleIO

[Martin Görner @martin_gorner]

Google I/O is today and "Large language models with Keras" is at 4:30PM 🔥with @smn_sdt and @GabrielRasskin !

[Gabriel Rasskin @GabrielRasskin]

Next-gen framework that vastly simplifies your JAX experience, and your Tensorflow, and your PyTorch. Somehow even your Numpy

Excited to see what OSS can build on top. Great effort from the team and community contributors

[François Chollet @fchollet]

Big news: we just released Keras 3.0!

▶ Run Keras on top of JAX, TensorFlow, and PyTorch
▶ Train faster with XLA compilation
▶ Unlock training runs with any number of devices & hosts via the new Keras distribution API

It's live on PyPI now! 🚀

[Gabriel Rasskin @GabrielRasskin]

@amasad As a heads up as long as you're using the latest version of express you can just do app.use(express.json()).

The AI is unfortunately wrong.

[Gabriel Rasskin @GabrielRasskin]

Small side project - easily create movie/TV themed titles at https://t.co/zPqaap2CiA

[Gabriel Rasskin @GabrielRasskin]

New #covidvisualizer sister site #vaccinevisualizer up and running; we're live tracking vaccine distribution globally: https://t.co/K9T9vvoIK1

[CMU School of Computer Science @SCSatCMU]

Get up-to-date information about the global COVID-19 epidemic with this interactive website, Covid Visualizer, developed by @CarnegieMellon students Navid Mamoon @CMU_DietrichHSS and Gabriel Rasskin @SCSatCMU https://t.co/AN8BJNILIZ