Google unveils updates to its Gemma family of open models, including Gemma 2 2B, which it claims surpasses GPT-3.5 and Mixtral 8x7B on the LMSYS Chatbot Arena
Google has unveiled updates to its Gemma 2 family of open-source language models, focusing on improved performance, safety, and transparency.
The Decoder Jonathan Kemper
Related Coverage
- Smaller, Safer, More Transparent: Advancing Responsible AI with Gemma Google Developers Blog
- View article Tech Monitor
- Google releases new ‘open’ AI models with a focus on safety TechCrunch · Kyle Wiggers
- View article Neowin
- View article SD Times
- View article Shelly Palmer
- Google's tiny AI model ‘Gemma 2 2B’ challenges tech giants in surprising upset VentureBeat · Michael Nuñez
- Google's new small AI model Gemma 2 2B surpasses OpenAI's GPT-3.5 The Indian Express
- Gemma 2-2B Released: A 2.6 Billion Parameter Model Offering Advanced Text Generation, On-Device Deployment, and Enhanced Safety Features MarkTechPost · Asif Razzaq
- Google DeepMind Introduces Gemma 2 AI Expansion With New Efficient Version WinBuzzer · Luke Jones
- Google's tiny AI beats GPT-3.5 The Rundown AI · Rowan Cheung
- Google's lightweight Gemma LLMs get smaller, but they perform even better than before SiliconANGLE · Mike Wheatley
- Google releases Gemma 2 2B, ShieldGemma and Gemma Scope Hugging Face · Xenova
- Google launches three ‘open’ AI models prioritizing safety and transparency NewsBytes · Mudit Dube
- Google Expands Gemma 2 Family with New 2B Model, Safety Classifiers, and Interpretability Tools Maginative · Chris McKay
- Google Unveils Gemma 2 2B, Outperforms GPT-3.5-Turbo and Mixtral-8x7B AIM · Siddharth Jindal
- Google DeepMind launches 2B parameter Gemma 2 model TNW · Ioanna Lykiardopoulou
- We're excited to introduce ShieldGemma, a groundbreaking suite of safety content classifier models designed to protect and elevate your AI interactions. … Wenjun Zeng
- Very proud to be launching our new family of AI models at Google DeepMind: — 🛡️ ShieldGemma is out! — Read the blog to learn more. … Ludovic Peran
Discussion
-
@lmsysorg
@lmsysorg
on x
Congrats @GoogleDeepMind on the Gemma-2-2B release! Gemma-2-2B has been tested in the Arena under “guava-chatbot”. With just 2B parameters, it achieves an impressive score 1130 on par with models 10x its size! (For reference: GPT-3.5-Turbo-0613: 1117, Mixtral-8x7b: 1114). This [i…
-
@googledeepmind
@googledeepmind
on x
We're also introducing ShieldGemma: a series of state-of-the-art safety classifiers designed to filter harmful content. 🛡️ These target hate speech, harassment, sexually explicit material and more, both in the input and output stages. [image]
-
@reach_vb
@reach_vb
on x
Google just dropped Gemma 2 2B! 🔥 > Scores higher than GPT 3.5, Mixtral 8x7B on the LYMSYS arena > MMLU: 56.1 & MBPP: 36.6 > Beats previous (Gemma 1 2B) by more than 10% in benchmarks > 2.6B parameters, Multilingual > 2 Trillion tokens (training set) > Distilled from Gemma 2 [ima…
-
@googledeepmind
@googledeepmind
on x
Finally, we're announcing Gemma Scope, a set of tools to help researchers examine how Gemma 2 makes decisions. 🔍 It's a comprehensive, open suite of sparse autoencoders - specialized neural networks that zoom into the model's inner workings and make them more interpretable. [imag…
-
@gblazex
Blaze
on x
The fact that we have a TINY model runnable on basically any smartphone + THIS closely aligned with human preferences & everyday usefulness is amazing. And all of it Open Weights permissive license incl. commercial use & fine-tuning. I mean come on... [image]
-
@chris_j_paxton
Chris Paxton
on x
It's old news now but GPT 3.5 - chatgpt - was the one that kicked off the whole LLM craze. to see it being beaten by a 2B paramater model is mind-blowing. This really means we can make every single thing we own (somewhat) intelligent [image]
-
@googledeepmind
@googledeepmind
on x
We're welcoming a new 2 billion parameter model to the Gemma 2 family. 🛠️ It offers best-in-class performance for its size and can run efficiently on a wide range of hardware. Developers can get started with 2B today → https://developers.googleblog.com/ ... [image]