# \#agent-platform-workbench

**URL:** https://discuss.google.dev/tag/agent-platform-workbench/390.md

[Latest](https://discuss.google.dev/latest.md) · [Categories](https://discuss.google.dev/categories.md) · [Tags](https://discuss.google.dev/tags.md)

---

## [Passing the GCP Professional Machine Learning Engineer (PMLE) in 1 Month: My Study Strategy & Key Resources](https://discuss.google.dev/t/passing-the-gcp-professional-machine-learning-engineer-pmle-in-1-month-my-study-strategy-key-resources/405272)

<div class="topic-metadata">

**Author:** [@snehachoudhary](https://discuss.google.dev/u/snehachoudhary)\
**Replies:** 0\
**Last updated:** [October 16, 2026, 7:29am UTC](https://discuss.google.dev/t/passing-the-gcp-professional-machine-learning-engineer-pmle-in-1-month-my-study-strategy-key-resources/405272 "2026-10-16T07:29:17Z")

</div>

I recently took the Google Cloud Professional Machine Learning Engineer (PMLE) exam. I cleared it in one month prep time. To be clear, I didn’t start from scratch. I already have experience with GCP and other cloud setu…

---

## [Graphs/plots not showing in Jupyter notebook in an instance on Vertex AI Workbench](https://discuss.google.dev/t/graphs-plots-not-showing-in-jupyter-notebook-in-an-instance-on-vertex-ai-workbench/287374)

<div class="topic-metadata">

**Author:** [@ignacy](https://discuss.google.dev/u/ignacy)\
**Replies:** 5\
**Last updated:** [December 4, 2025, 2:46am UTC](https://discuss.google.dev/t/graphs-plots-not-showing-in-jupyter-notebook-in-an-instance-on-vertex-ai-workbench/287374 "2025-12-04T02:46:23Z")

</div>

Hello! I have bumped into an initially innoncent issue, that spiralled into an actual problem that I have yet to solve. To put it simply, none of my instances I have deployed in Vertex AI Workbench are currently capable…

---

## [\[Vertex AI Bug\] Project cannot access ANY Gemini models ("Listed 0 items") - Extensive Troubleshooting Done](https://discuss.google.dev/t/vertex-ai-bug-project-cannot-access-any-gemini-models-listed-0-items-extensive-troubleshooting-done/275270)

<div class="topic-metadata">

**Author:** [@75\_020](https://discuss.google.dev/u/75_020)\
**Replies:** 1\
**Last updated:** [October 20, 2025, 8:40am UTC](https://discuss.google.dev/t/vertex-ai-bug-project-cannot-access-any-gemini-models-listed-0-items-extensive-troubleshooting-done/275270 "2025-10-20T08:40:20Z")

</div>

Hello everyone, Google team, Hoping someone can help me out here because this is driving me crazy. I’m completely blocked from using any Gemini models through Vertex AI on my project, and I’ve tried pretty much everythi…

---

## [Adk Agent Deployment Not working](https://discuss.google.dev/t/adk-agent-deployment-not-working/191030)

<div class="topic-metadata">

**Author:** [@Papoulas](https://discuss.google.dev/u/Papoulas)\
**Replies:** 9\
**Last updated:** [September 30, 2025, 2:45pm UTC](https://discuss.google.dev/t/adk-agent-deployment-not-working/191030 "2025-09-30T14:45:28Z")

</div>

I have an application built with Google’s new Agent Development Kit (ADK) that I’ve successfully deployed to Vertex AI Agent Engine in GCP. The deployment completes without errors, and I can confirm in the GCP console th…

---

## [Can we please get an offline token-counter so RAG chunkers can work reliably w/ embedding-exp-03-07](https://discuss.google.dev/t/can-we-please-get-an-offline-token-counter-so-rag-chunkers-can-work-reliably-w-embedding-exp-03-07/184624)

<div class="topic-metadata">

**Author:** [@cndg](https://discuss.google.dev/u/cndg)\
**Replies:** 4\
**Last updated:** [July 25, 2025, 2:54pm UTC](https://discuss.google.dev/t/can-we-please-get-an-offline-token-counter-so-rag-chunkers-can-work-reliably-w-embedding-exp-03-07/184624 "2025-07-25T14:54:14Z")

</div>

I’m quite proud of my “chunker” for my custom RAG, which has some elegant recursive mechanisms based on tiktoken: tokenizer = tiktoken.encoding\_for\_model("text-embedding-3-large") I’m wanting to upgrade to gemini-embed…

---

## [MALFORMED\_FUNCTION\_CALL finish reason happens too frequently with gemini from vertex AI](https://discuss.google.dev/t/malformed-function-call-finish-reason-happens-too-frequently-with-gemini-from-vertex-ai/194090)

<div class="topic-metadata">

**Author:** [@darboxx](https://discuss.google.dev/u/darboxx)\
**Replies:** 2\
**Last updated:** [July 14, 2025, 5:14pm UTC](https://discuss.google.dev/t/malformed-function-call-finish-reason-happens-too-frequently-with-gemini-from-vertex-ai/194090 "2025-07-14T17:14:54Z")

</div>

I often get the MALFORMED\_FUNCTION\_CALL finish reason. This is issue is extremely frequent with Vertex AI and less frequent with Google AI developer API. To repro, use this import google.genai as genai from dotenv impor…

---

## [Unable to create a Vertex AI workspace instance. Nvidia T4 GPU Resource unavailable](https://discuss.google.dev/t/unable-to-create-a-vertex-ai-workspace-instance-nvidia-t4-gpu-resource-unavailable/193121)

<div class="topic-metadata">

**Author:** [@sandeepnjk](https://discuss.google.dev/u/sandeepnjk)\
**Replies:** 1\
**Last updated:** [July 2, 2025, 6:55pm UTC](https://discuss.google.dev/t/unable-to-create-a-vertex-ai-workspace-instance-nvidia-t4-gpu-resource-unavailable/193121 "2025-07-02T18:55:01Z")

</div>

Its been 2 days i am trying to create a Vertex AI workspace instance with Nvidia T4 GPU attached. But no luck. I have checked the quotas, but still no success. Without the nvidia GPU T4 selected i was able to create a in…

---

## [Persistent 404 Error Accessing Any Vertex AI Model Across Multiple Projects](https://discuss.google.dev/t/persistent-404-error-accessing-any-vertex-ai-model-across-multiple-projects/192780)

<div class="topic-metadata">

**Author:** [@Lenilson](https://discuss.google.dev/u/Lenilson)\
**Replies:** 2\
**Last updated:** [July 1, 2025, 6:17pm UTC](https://discuss.google.dev/t/persistent-404-error-accessing-any-vertex-ai-model-across-multiple-projects/192780 "2025-07-01T18:17:37Z")

</div>

Hello everyone, I’m facing a persistent and perplexing issue where I am unable to make any successful API calls to Vertex AI’s generative models from my account. 1. Detailed Issue: Every attempt to call model.generate…

---

## [Google vertex api quota](https://discuss.google.dev/t/google-vertex-api-quota/192696)

<div class="topic-metadata">

**Author:** [@mk1998](https://discuss.google.dev/u/mk1998)\
**Replies:** 0\
**Last updated:** [June 25, 2025, 10:30pm UTC](https://discuss.google.dev/t/google-vertex-api-quota/192696 "2025-06-25T22:30:53Z")

</div>

I have tried to submit several tickets pertaining to my google vertex api quota . i have requested a increased with no success and have been told multiple times to contact a partner company or to contact sales for a mana…

---

## [Error HttpResponse for each new AI Agent deployed to AI Engine](https://discuss.google.dev/t/error-httpresponse-for-each-new-ai-agent-deployed-to-ai-engine/192118)

<div class="topic-metadata">

**Author:** [@Alekzan](https://discuss.google.dev/u/Alekzan)\
**Replies:** 1\
**Last updated:** [June 25, 2025, 5:48pm UTC](https://discuss.google.dev/t/error-httpresponse-for-each-new-ai-agent-deployed-to-ai-engine/192118 "2025-06-25T17:48:18Z")

</div>

I have deployed an Agent to Vertex AI following the documentation: # DEPLOYMENT: Deploy the app to Vertex AI from vertexai import agent\_engines from orchestrator\_agent import root\_agent remote\_app = agent\_engines.creat…

---

## [Internal error occurred on vertex autoML training](https://discuss.google.dev/t/internal-error-occurred-on-vertex-automl-training/191935)

<div class="topic-metadata">

**Author:** [@jphilippi\_trust](https://discuss.google.dev/u/jphilippi_trust)\
**Replies:** 3\
**Last updated:** [June 20, 2025, 6:12am UTC](https://discuss.google.dev/t/internal-error-occurred-on-vertex-automl-training/191935 "2025-06-20T06:12:02Z")

</div>

the vertex autoML training process send me the message: " Training pipeline failed with error message: Internal error occurred. Please retry in a few minutes. If you still experience errors, contact Vertex AI." when the…

---

## [Error: ResourceExhausted: 429 Quota exceeded for aiplatform.googleapis.com/online\_prediction\_request](https://discuss.google.dev/t/error-resourceexhausted-429-quota-exceeded-for-aiplatform-googleapis-com-online-prediction-request/127161)

<div class="topic-metadata">

**Author:** [@KannanG03](https://discuss.google.dev/u/KannanG03)\
**Replies:** 10\
**Last updated:** [June 16, 2025, 5:16am UTC](https://discuss.google.dev/t/error-resourceexhausted-429-quota-exceeded-for-aiplatform-googleapis-com-online-prediction-request/127161 "2025-06-16T05:16:45Z")

</div>

Is there any limitation on generating embeddings? I tried to generate embeddings for 152 chunks, with each chunk containing only one array element. However, I am still encountering the ‘429 Quota Exceeded’ error. I have …

---

## [Signed NVIDIA drivers for Debian](https://discuss.google.dev/t/signed-nvidia-drivers-for-debian/191251)

<div class="topic-metadata">

**Author:** [@leesmith](https://discuss.google.dev/u/leesmith)\
**Replies:** 1\
**Last updated:** [June 13, 2025, 1:27pm UTC](https://discuss.google.dev/t/signed-nvidia-drivers-for-debian/191251 "2025-06-13T13:27:10Z")

</div>

Hello! My organization requires all of our instances run with secure boot enabled. The consequence of this is that we require google signed drivers. At the moment vertex recommends using debian images and is deprecating…

---

## [Internal Server Error (Code 13) with Vertex AI Auto ML Image Classification (Single Label)](https://discuss.google.dev/t/internal-server-error-code-13-with-vertex-ai-auto-ml-image-classification-single-label/191485)

<div class="topic-metadata">

**Author:** [@harsh088](https://discuss.google.dev/u/harsh088)\
**Replies:** 0\
**Last updated:** [June 12, 2025, 7:45am UTC](https://discuss.google.dev/t/internal-server-error-code-13-with-vertex-ai-auto-ml-image-classification-single-label/191485 "2025-06-12T07:45:41Z")

</div>

Hello Google Cloud Community, I’m encountering a persistent “Internal Server Error” (Code 13) when attempting to train a single-label image classification model using Vertex AI AutoML. I’m following the official Google …

---

## [Vertext AI → Media Studio → Imagen 4: Cannot use images as part of a prompt?](https://discuss.google.dev/t/vertext-ai-media-studio-imagen-4-cannot-use-images-as-part-of-a-prompt/190395)

<div class="topic-metadata">

**Author:** [@ylluminate](https://discuss.google.dev/u/ylluminate)\
**Replies:** 2\
**Last updated:** [June 3, 2025, 3:18pm UTC](https://discuss.google.dev/t/vertext-ai-media-studio-imagen-4-cannot-use-images-as-part-of-a-prompt/190395 "2025-06-03T15:18:02Z")

</div>

When I attempt to use Imagen 4, I expect to be able to upload an image to use as part of a prompt as is done in ChatGPT and even other permutations of Google’s AI image generation tooling. When I upload an image into the…

---

## [403 when running Workbench Instance](https://discuss.google.dev/t/403-when-running-workbench-instance/189830)

<div class="topic-metadata">

**Author:** [@julliet](https://discuss.google.dev/u/julliet)\
**Replies:** 1\
**Last updated:** [May 28, 2025, 2:02pm UTC](https://discuss.google.dev/t/403-when-running-workbench-instance/189830 "2025-05-28T14:02:29Z")

</div>

I was running User-managed notebooks before with the same settings having no issue. Now I’m running new instance and want to use Jupyter Notebook UI. I’m running instance from service account. For fast troubleshooting…

---

## [Vertex AI Managed Notebooks conflicting dependencies issue.](https://discuss.google.dev/t/vertex-ai-managed-notebooks-conflicting-dependencies-issue/188891)

<div class="topic-metadata">

**Author:** [@Archa](https://discuss.google.dev/u/Archa)\
**Replies:** 6\
**Last updated:** [May 15, 2025, 5:07pm UTC](https://discuss.google.dev/t/vertex-ai-managed-notebooks-conflicting-dependencies-issue/188891 "2025-05-15T17:07:05Z")

</div>

Hi, Recently I tried to install a new package in my managed notbook instance of Vertex AI workbench. But I got an error stating that jupyterlab failed to open due to conflicting dependencies after the new package instal…

---

## [Cannot create runtime for Colab despite having quota, cannot sign up for technical support.](https://discuss.google.dev/t/cannot-create-runtime-for-colab-despite-having-quota-cannot-sign-up-for-technical-support/189268)

<div class="topic-metadata">

**Author:** [@BecomingPhill](https://discuss.google.dev/u/BecomingPhill)\
**Replies:** 0\
**Last updated:** [May 15, 2025, 8:00am UTC](https://discuss.google.dev/t/cannot-create-runtime-for-colab-despite-having-quota-cannot-sign-up-for-technical-support/189268 "2025-05-15T08:00:31Z")

</div>

I have been using Colab Pro Plus to run phsyics and cosmology simulations on an A100, I am using my Google Drive to store snapshot and analysis data. I require access to large GPUs to run the simulation in a more effici…

---

## [Quota cycle for GPUs](https://discuss.google.dev/t/quota-cycle-for-gpus/188011)

<div class="topic-metadata">

**Author:** [@danherman212245](https://discuss.google.dev/u/danherman212245)\
**Replies:** 1\
**Last updated:** [April 30, 2025, 12:46am UTC](https://discuss.google.dev/t/quota-cycle-for-gpus/188011 "2025-04-30T00:46:26Z")

</div>

I was using a GPU for training and got past the soft quota of 1 GPU I deleted my workbench instance, but the quota still shows that I have 100% utilization, 3 hrs later Can anyone provide guidance on how long it takes …

---

## [Error 403 when accessing JupyterLab in Vertex AI Workbench instance](https://discuss.google.dev/t/error-403-when-accessing-jupyterlab-in-vertex-ai-workbench-instance/187364)

<div class="topic-metadata">

**Author:** [@micaelbatista](https://discuss.google.dev/u/micaelbatista)\
**Replies:** 2\
**Last updated:** [April 28, 2025, 8:58am UTC](https://discuss.google.dev/t/error-403-when-accessing-jupyterlab-in-vertex-ai-workbench-instance/187364 "2025-04-28T08:58:31Z")

</div>

I’m trying to create a new instance of a JupyterLab in Vertex AI Workbench, but when I click in “Open JupyterLab” I always get a 403 back. I already have the Vertex AI Administrator role, but since I’m getting this issue…

---

## [VertexAISearch (grounding from data store). 400 Request contains an invalid argument.](https://discuss.google.dev/t/vertexaisearch-grounding-from-data-store-400-request-contains-an-invalid-argument/160264)

<div class="topic-metadata">

**Author:** [@msdavid](https://discuss.google.dev/u/msdavid)\
**Replies:** 8\
**Last updated:** [April 24, 2025, 4:38am UTC](https://discuss.google.dev/t/vertexaisearch-grounding-from-data-store-400-request-contains-an-invalid-argument/160264 "2025-04-24T04:38:32Z")

</div>

I am trying to use the Vertex AI search grounding. I followed this notebook. I keep getting a InvalidArgument: 400 Request contains an invalid argument. this is my stripped down code: import vertexai from vertexai.gen…

---

## [FastAPI + Docker Deployment Fails on Vertex AI with 404 Error (GET /v1/endpoints/.../deployedModels/](https://discuss.google.dev/t/fastapi-docker-deployment-fails-on-vertex-ai-with-404-error-get-v1-endpoints-deployedmodels/187282)

<div class="topic-metadata">

**Author:** [@suryaprakash\_b](https://discuss.google.dev/u/suryaprakash_b)\
**Replies:** 1\
**Last updated:** [April 23, 2025, 12:05pm UTC](https://discuss.google.dev/t/fastapi-docker-deployment-fails-on-vertex-ai-with-404-error-get-v1-endpoints-deployedmodels/187282 "2025-04-23T12:05:58Z")

</div>

Hi, I’m trying to deploy a FastAPI project using a Dockerfile to Vertex AI via gcloud CLI (models upload, endpoints create, deploy-model). I’ve successfully uploaded the model and created the endpoint. However, the depl…

---

## [Unable to Schedule Vertex AI Workbench notebooks using a custom container](https://discuss.google.dev/t/unable-to-schedule-vertex-ai-workbench-notebooks-using-a-custom-container/186543)

<div class="topic-metadata">

**Author:** [@dfry\_lhzn\_io](https://discuss.google.dev/u/dfry_lhzn_io)\
**Replies:** 0\
**Last updated:** [April 11, 2025, 3:38pm UTC](https://discuss.google.dev/t/unable-to-schedule-vertex-ai-workbench-notebooks-using-a-custom-container/186543 "2025-04-11T15:38:47Z")

</div>

Hi! I would like to schedule notebook execution using a custom container with a custom default kernel and a custom least-privilege service account. It seems the standard base container excludes nbexecutor. I attempted t…

---

## [Error: "Project '12345' is not allowed to use Publisher model"](https://discuss.google.dev/t/error-project-12345-is-not-allowed-to-use-publisher-model/186538)

<div class="topic-metadata">

**Author:** [@sbarnhill](https://discuss.google.dev/u/sbarnhill)\
**Replies:** 0\
**Last updated:** [April 11, 2025, 2:40pm UTC](https://discuss.google.dev/t/error-project-12345-is-not-allowed-to-use-publisher-model/186538 "2025-04-11T14:40:57Z")

</div>

During the “Compare Model Performance using the Generative AI Evaluation Service: Challenge Lab” I am running into a technical access issue with the Gemini 1.5 Pro and Gemini 1.5 Flash models. I am not looking for the co…

---

## [How do you add "parameters" to a batch prediction job's POST request JSON body?](https://discuss.google.dev/t/how-do-you-add-parameters-to-a-batch-prediction-jobs-post-request-json-body/127685)

<div class="topic-metadata">

**Author:** [@anonymous8](https://discuss.google.dev/u/anonymous8)\
**Replies:** 4\
**Last updated:** [April 9, 2025, 1:53am UTC](https://discuss.google.dev/t/how-do-you-add-parameters-to-a-batch-prediction-jobs-post-request-json-body/127685 "2025-04-09T01:53:51Z")

</div>

How do you configure a BatchPredictionJob to include a “parameters” JSON object (i.e., a dict) in the JSON body of each POST request? I’ve configured a custom-container based Model to expect “parameters” in the body of a…

---

## [No Module Named bq\_stats](https://discuss.google.dev/t/no-module-named-bq-stats/185441)

<div class="topic-metadata">

**Author:** [@mplb](https://discuss.google.dev/u/mplb)\
**Replies:** 3\
**Last updated:** [April 4, 2025, 9:10am UTC](https://discuss.google.dev/t/no-module-named-bq-stats/185441 "2025-04-04T09:10:32Z")

</div>

I’m using the new instances of the vertex notebooks. When attempting to run the scheduled jobs I get this message: I’ve tried pip installing that module but it can’t be found/says it’s already installed.

---

## [Error predicting from a vertex ai endpoint in python (error 503 DNS resolution failed)](https://discuss.google.dev/t/error-predicting-from-a-vertex-ai-endpoint-in-python-error-503-dns-resolution-failed/185459)

<div class="topic-metadata">

**Author:** [@ducc](https://discuss.google.dev/u/ducc)\
**Replies:** 1\
**Last updated:** [April 2, 2025, 5:35pm UTC](https://discuss.google.dev/t/error-predicting-from-a-vertex-ai-endpoint-in-python-error-503-dns-resolution-failed/185459 "2025-04-02T17:35:16Z")

</div>

Traceback (most recent call last): File "c:\\Users\\zacis\\AppData\\Local\\Programs\\Python\\Python311\\Lib\\site-packages\\google\\api\_core\\grpc\_helpers.py", line 76, in error\_remapped\_callable return callable\_(\*args, \*\*kwar…

---

## [Vector Search Quickstart Error](https://discuss.google.dev/t/vector-search-quickstart-error/185165)

<div class="topic-metadata">

**Author:** [@JackJonesTrellx](https://discuss.google.dev/u/JackJonesTrellx)\
**Replies:** 2\
**Last updated:** [March 31, 2025, 6:20pm UTC](https://discuss.google.dev/t/vector-search-quickstart-error/185165 "2025-03-31T18:20:35Z")

</div>

I am following the Vector Search Quickstart Guide, https://cloud.google.com/vertex-ai/docs/vector-search/quickstart#install-sdk, and encountering an issue when trying to run the following query in Jupyter notebook: run …

---

## [Trying to build document search engine using vertexAI getting very large response](https://discuss.google.dev/t/trying-to-build-document-search-engine-using-vertexai-getting-very-large-response/184423)

<div class="topic-metadata">

**Author:** [@kavanadeshpande](https://discuss.google.dev/u/kavanadeshpande)\
**Replies:** 1\
**Last updated:** [March 21, 2025, 1:32pm UTC](https://discuss.google.dev/t/trying-to-build-document-search-engine-using-vertexai-getting-very-large-response/184423 "2025-03-21T13:32:00Z")

</div>

I’m working on building document search engine using vertexAI. I have created search app and it’s knowledge base has all documents required. I’m building python wrapper for the same using discoveryengine.ConversationalSe…

---

## [Error "failed to list the clusters" creating cloud serverless runtime template](https://discuss.google.dev/t/error-failed-to-list-the-clusters-creating-cloud-serverless-runtime-template/151010)

<div class="topic-metadata">

**Author:** [@jochen\_hegele](https://discuss.google.dev/u/jochen_hegele)\
**Replies:** 5\
**Last updated:** [March 20, 2025, 8:09am UTC](https://discuss.google.dev/t/error-failed-to-list-the-clusters-creating-cloud-serverless-runtime-template/151010 "2025-03-20T08:09:32Z")

</div>

Hello, I try to create a dataproc cloud serverless runtime template to get a spark environment notebook in a vertex ai workbench instance. If click on the “New Runtime Template” in category Dataproc Serverless Notebook…

[Next page](https://discuss.google.dev/tag/agent-platform-workbench/390.md?match_all_tags=true&page=1&tags%5B%5D=agent-platform-workbench)
