Google Gemini / Vertex AI

Vertex AI Generative AI Update: RAG Engine, Veo 3.1 Lite, and Endpoint Deprecations

Generative AI on Vertex AI v1 Change RAG Engine is generally available (GA) . The supported models include the following: The supported features include the following: Data connectors : Google Cloud Storage, Google Drive, Slack, Jira, and SharePoint Document types : Google Workspace documents, HTML, JSON, Markdown, PDF, and text files Transformations : fixed-size chunking and chunk overlap Vector databases : Vertex AI Vector Search and Pinecone Vertex AI v1 Feature Vector Search hybrid search and sparse embeddings are generally available (GA) Vector Search hybrid search and sparse embeddings are generally available ( GA ). Hybrid search uses both dense and sparse embeddings, which lets you search based on a combination of keyword search and semantic search. For more information about hybrid search, see About hybrid search .

Date not specified

Vertex AI Workbench M127 Release: SSH Key Fix & Image Migration

Vertex AI Workbench v1 Feature M127 release The M127 release of Vertex AI Workbench user-managed notebooks includes the following: Fixed an issue related to ownership of the home directory when using authorized ssh keys. Feature The M127 release of Vertex AI Workbench managed notebooks includes the following: Fixed an issue related to ownership of the home directory when using authorized ssh keys. v2 Feature M127 release The M127 release of Vertex AI Workbench instances includes the following: Fixed an issue related to ownership of the home directory when using authorized ssh keys.

Date not specified

Vertex AI LangChain Workloads Pricing Changes Starting March 4, 2025

Generative AI on Vertex AI v1 Announcement LangChain on Vertex AI Billing for LangChain on Vertex AI will start on March 4, 2025. The pricing structure is based on vCPU hours and GiB hours used. This means that you will be charged for both the compute (vCPU) and memory resources consumed by your LangChain on Vertex AI workloads. You can review the pricing details in the table below. Product SKU ID Price ReasoningEngine vCPU 8A55-0B95-B7DC $0.0994/vCPU-Hr ReasoningEngine Memory 0B45-6103-6EC1 $0.0105/GiB-Hr Vertex AI Agent Builder Announcement LangChain on Vertex AI Billing for LangChain on Vertex AI will start on March 4, 2025. The pricing structure is based on vCPU hours and GiB hours used. This means that you will be charged for both the compute (vCPU) and memory resources consumed by your LangChain on Vertex AI workloads. You can review the pricing details in the table below. Product SKU ID Price ReasoningEngine vCPU 8A55-0B95-B7DC $0.0994/vCPU-Hr ReasoningEngine Memory 0B45-6103-6EC1 $0.0105/GiB-Hr

Date not specified

Google releases Imagen 3 image generation model on Vertex AI

Generative AI on Vertex AI v1 Feature New Imagen 3 image generation model available to users A newer improved Imagen 3 image generation model is now available to all users: This image generation model supports the following additional features: Prompt enhancement - The LLM-based prompt rewriter tool adds additional details and descriptive language to the prompt you provide, generally resulting in higher quality generated images. This feature is configurable and is enabled by default. For more information, see Imagen on Vertex AI model versions and lifecycle and Generate images using text prompts .

Date not specified

Vertex AI releases new monitoring dashboard for foundation models

Generative AI on Vertex AI v1 Feature You can now monitor usage, throughput, and latency and troubleshoot 429 errors on Vertex AI foundation models, like Google Gemini and Anthropic Claude, by using a predefined dashboard. After querying a model from the Vertex AI Model Garden , you can find the name of the model you queried in the Vertex AI Dashboard page under the "Model observability" heading. To customize the dashboard and explore relevant metrics in Cloud Monitoring, click Show All Metrics . For information about using dashboards in Cloud Monitoring, see View and customize Google Cloud dashboards . Vertex AI v1

Date not specified

Google Releases Gemini 2.0 Models & Deprecates Imagen/Veo Generation Endpoints

Generative AI on Vertex AI v1 Feature Gemini 2.0 Flash general availability for text-only output Gemini 2.0 Flash is now generally available for text-only outputs. Multimodal outputs are still available only as a private preview. For more information, see Gemini 2.0 . Feature New Gemini 2.0 Pro and Gemini 2.0 Flash-Lite models available to users Two new models in the Gemini 2.0 family are now available to users: Gemini 2.0 Pro: Our strongest model for coding and world knowledge, featuring a 2M long context window. Gemini 2.0 Pro is available as an experimental model in Vertex AI. Gemini 2.0 Flash-Lite: Our fastest and most cost efficient Flash model. Gemini 2.0 Flash-Lite is available as a Preview model in Vertex AI. For more information, see Gemini 2.0

Date not specified

Vertex AI Model Garden: New LLM Inference Optimization Techniques Released

Generative AI on Vertex AI v1 Change The following advanced LLM inference optimization techniques are available in Model Garden in Preview : Prefix caching reuses computations from previously generated text, eliminating redundant processing. It reduces time-to-first-token for requests with common prompt prefixes. Prefix caching is available for the following models: vLLM: Llama 3.1 (8b, 70b), Llama 3.3 (70b) Hex-LLM: Llama 2 (7b, 13b), Llama 3 (8b), Llama 3.1 (8b, 70b), Llama 3.2 (1b, 3b), Llama Guard (1b, 8b), CodeLlama (7b, 13b), Gemma (2b, 7b), CodeGemma (2b, 7b), Mistral-7B (v0.2, v0.3), Mixtral-8x7B (v0.1) Speculative decoding is an effective optimization technique to reduce generation time-per-output-token latency. For more information, see the Model Garden advanced features notebook .

Date not specified

Google Adds DeepSeek Models to Vertex AI's Model Garden

Generative AI on Vertex AI v1 Feature Deepseek-V3 and Deepseek-R1 have been added to Model Garden in Preview : DeepSeek-V3 (671B) is a powerful Mixture-of-Experts (MoE) language model with 671B total parameters with 37B activated for each token. DeepSeek-R1 (671B) is one of the first-generation reasoning models introduced by DeepSeek and offers performance comparable to OpenAI-o1 across math, code, and reasoning tasks. You can use a notebook to deploy these models.

Date not specified

Vertex AI Generative AI Updates: Model Releases, Feature Enhancements, and Deprecations

Generative AI on Vertex AI v1 Change PEFT Docker updates Added support for evaluation metrics like perplexity, bleu, google_bleu, rouge1, rouge2, rougeL, rougeLSum. Uses the best checkpoint and loads the model based on the best eval metrics. Run training and eval only for data which is less than or equal to the max_seq_length . Use gcloud storage rsync instead of csfuse to save a checkpoint. Fine tuning updates Model updates Updated the PaliGemma model card by supporting PaliGemma 2 mix models, and segmentation functionality to Paligemma 1 models. Updated the LLaVa model card by supporting LLaVA Next models and adding vLLM to the notebook.

Date not specified

Colab Enterprise: Terraform Support for Notebook Scheduling

Colab Enterprise Feature You can use Terraform resources to schedule notebook runs, and to manage runtimes and runtime templates. To learn more, see the following:

Date not specified

Vertex AI Generative AI Update: Gemini 2.0 Flash Tuning & Deprecations

Generative AI on Vertex AI v1 Change Gemini 2.0 Flash Tuning Gemini 2.0 Flash fine-tuning is now generally available ( GA ). Added support for tuning function calling .

Date not specified

Vertex AI Updates: M128 Workbench Release, New Models, and Deprecations

Generative AI on Vertex AI v1 Change Model Garden fine tuning updates: Vertex AI Workbench v1 Feature M128 release The M128 release of Vertex AI Workbench user-managed notebooks includes the following: Miscellaneous package updates. Feature The M128 release of Vertex AI Workbench managed notebooks includes the following: Miscellaneous package updates. v2 Feature M128 release The M128 release of Vertex AI Workbench instances includes the following: Miscellaneous package updates.

Date not specified

Google releases Mistral Small 3.1 with multimodal capabilities on Vertex AI

Generative AI on Vertex AI v1 Announcement Mistral Small 3.1 (25.03) feature multimodal capabilities and a context of up to 128,000 tokens. For more information, see the Mistral Small 3.1 (25.03) model card in Model Garden.

Date not specified

Vertex AI Updates: New Models & Deprecations - March 20, 2025

Generative AI on Vertex AI v1 Vertex AI Workbench v2

Date not specified

Vertex AI v1 Release - New Features, Model Updates, and Deprecations

Vertex AI v1 Feature Generally available : You can consume reservations of VMs that have GPUs attached with your custom training jobs or prediction jobs. Reservations of Compute Engine zonal resources help you gain a high level of assurance that your jobs have the necessary resources to run. For more information, see the following: Vertex AI Workbench v2

Date not specified

Vertex AI introduces Spot VMs for cost savings

Vertex AI v1 Feature Generally available : To reduce the cost of running your training and prediction jobs, you can use Spot VMs. Spot VMs are virtual machine (VM) instances that are excess Compute Engine capacity. Spot VMs have significant discounts, but Compute Engine might preemptively stop or delete Spot VMs to reclaim the capacity at any time. For more information, see Use Spot VMs with training and Use Spot VMs with prediction .

Date not specified

Google Colab Enterprise: Default GPU Runtime Feature Preview

Colab Enterprise Feature Preview : You can switch to a default runtime with GPUs by using a button in your Colab Enterprise notebook. To enable a default runtime with GPUs for your users, see Enable default runtimes with GPUs .

Date not specified

Vertex AI Updates: New Features, Model Releases, and Deprecations

Generative AI on Vertex AI v1 Feature Vertex AI Agent Engine The following features are now available for Vertex AI Agent Engine in Preview: The following features are now generally available for Vertex AI Agent Engine: Feature Gemini Live API is now available as a public preview offering and has been updated with the following features: Support for responses in 8 voices and 31 languages using Chirp 3 Updated UI support in Vertex AI Studio Expanded conversation session window Ability to extend conversation sessions Support to share your current screen with Gemini during conversations Transcription support for audio in and audio out Support to change or update the system instructions mid-session For more information, see Gemini 2.0 Flash Live API . Feature Gemini 2.5 Pro is now available as a public preview offering. For more information, see Gemini 2.5 Pro . Change Vertex AI Agent Builder now refers to a suite of features for building and deploying AI agents in Vertex AI. For more information see, Vertex AI Agent Builder overview . The original Vertex AI Agent Builder product has been renamed AI Applications . The product functionality and endpoints remain the same. For more information, see What is AI Applications? . Feature Grounding: Grounding with Google Maps is now available as a Public Experimental feature. For more information, see Grounding with Google Maps . Vertex AI Agent Builder Feature Vertex AI Agent Engine The following features are now available for Vertex AI Agent Engine in Preview: The following features are now generally available for Vertex AI Agent Engine: Change Vertex AI Agent Builder now refers to a suite of features for building and deploying AI agents in Vertex AI. For more information see, Vertex AI Agent Builder overview . The original Vertex AI Agent Builder product has been renamed AI Applications . The product functionality and endpoints remain the same. For more information, see What is AI Applications? .

Date not specified

Vertex AI Workbench v2 receives M129 update with Debian 12 migration

Vertex AI v1 Vertex AI Workbench v2 Feature M129 release The M129 release of Vertex AI Workbench instances includes the following: Updated the Dataproc JupyterLab plugin to version 0.1.85.

Date not specified

Google Colab Enterprise Launches Notebook Gallery Feature

Colab Enterprise Feature The notebook gallery is now available. The notebook gallery is a curated collection of notebooks to help you get started using Colab Enterprise. This collection consists of ready-to-use templates and examples to make it easier to learn new techniques, understand best practices, and get projects started quickly. Browse the notebooks by category or use the search bar to find a notebook that helps you get started. See the notebook gallery .

Date not specified