Google says Gemma 2 will be available to researchers and developers through Vertex AI from July as a new 9B-parameter model and the prior 27B-parameter model
Putting the 9B and 27B Gemma 2 models on Vertex AI matters because it joins model availability to Google’s managed AI platform, rather than leaving developers to treat access to the model and deployment infrastructure as separate choices.
First-order effects
Researchers and developers using Vertex AI gain access to Gemma 2 in a new 9B size and the 27B model from July, widening the model-size options available through Google’s platform.
Google makes Vertex AI a direct distribution channel for Gemma 2, connecting its model family more closely to its managed development environment.
Second-order effects
Teams choosing between the two sizes can align model selection with their workload and deployment requirements within one platform, reducing the need to assemble separate model-access and serving workflows.
The move increases the value of Vertex AI’s model catalog: Google’s specialized Gemma releases and managed access can reinforce each other for developers already evaluating Google’s AI stack.
Third-order effects
If this pattern continues, open or broadly usable model families will increasingly function as on-ramps to cloud AI platforms, making deployment tooling and inference operations a larger source of differentiation than model access alone.
The Gemma roadmap points toward a more segmented model market, where size and workload-specific variants give buyers more choice but also make platform-level integration central to adoption.
The trend: Cloud providers are pairing increasingly differentiated model families with managed platforms to turn model choice into longer-term infrastructure adoption.
Google's new Gemini and Vertex AI announcements gives enterprise customers more flexibility when building apps. The first: Gemini 1.5 Flash + 1.5 Pro w/2M token context window now GA. Plus, they're getting context caching and provisioned throughout. https://venturebeat.com/...
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 […
Lots of big announcements today! Make sure to check out this intro to what you can do with Gemma 2: https://developers.googleblog.com/ ... Gemma 2 27B is pretty insane — a GPT-4 level open model about 10x smaller than what you'd expect. You can literally run it in Colab.
Gemma 2 is available to researchers & developers. At 27B it delivers best-in-class performance for its size and is competitive even to models over twice its size! Proud to continue our tradition of thoughtfully bringing cutting-edge research to the open models ecosystem.
Gemma 27B Is Now The Top Open Source Model On LymSys?! I was reading this wrong in the morning, but it seems like Gemma 27B is the top open-source model. Human eval arena IMO, this is a highly gamed benchmark, and we will know a lot more about this model once we have the [image]
Congrats @GoogleDeepMind on the new Gemma-2 27B & 9B release! Gemma-2 was tested in the Arena under the codename “*late-june-chatbots” and now out of stealth. Its early result matches the best open models (Llama-3-70B, Nemotron-340B) with only 27B parameters! Impressively, [image…
The 2M context window is great to dump books and then query the book. Also, I pasted terrible PDF cut-and-paste that was close to impossible to read with any comfort and it was still great.
Let's fucking gooo! Google just dropped Gemma 2 27B & 9B 🔥 > Beats Llama3 70B/ Qwen 72B/ Command R+ in LYMSYS Chat arena & 9B is the best < 15B model right now. > 2.5x smaller than Llama 3 & trained on 2/3rd the amount of tokens > Trained on 13T tokens (27B) and 8T (tokens) > [im…
More great news for @GoogleAI developers: - Gemini 1.5 Pro 2M context is available to all - Code execution tool is now available with Gemini 1.5 - Gemini 1.5 Flash text tuning has started gradually rolling out - Gemma 2 is now available in AI Studio https://developers.googleblog.…
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://www.kaggle.com/... On…
Gemma 2 is out! As with our first model, we're super focused on creating models at useful, practical sizes, so that they can be easily deployable... all the while being amazing in quality. We upgraded our 9B so that it's truly awesome and best in class across many benchmarks. [im…
Gemma 2 is officially here! 🥳 Available in both 9B and 27B parameter sizes, Gemma 2 is higher performing than ever before. Download the weights on @Kaggle and @HuggingFace, or access the models in Google AI Studio. Learn more → https://blog.google/... [video]
I am so proud to announce that: - Gemma 2 27B IT tops all open weights models on Chatbot Arena, with a pinch of optimism in the face of uncertainty :) - Gemma 2 9B IT sets a new frontier for models of similar size. 1/n [image]