cancel
Showing results for 
Search instead for 
Did you mean: 
Generative AI
Explore discussions on generative artificial intelligence techniques and applications within the Databricks Community. Share ideas, challenges, and breakthroughs in this cutting-edge field.
cancel
Showing results for 
Search instead for 
Did you mean: 

Forum Posts

Ale_Armillotta
by Valued Contributor II
  • 291 Views
  • 2 replies
  • 0 kudos

MLFlow tracking from Azure Container Instance

Hi everyone,we're building a voice chatbot for a customer using a mix of technologies — Databricks, Azure AI Foundry, and a few external containerized services.Currently, we're tracking requests and logs via Lakebase with custom traces, but I'm now e...

  • 291 Views
  • 2 replies
  • 0 kudos
Latest Reply
MoJaMa
Databricks Employee
  • 0 kudos

Similar response to the one from WorksBuddy.   Short answer: yes, it's supported, and there's a specific Databricks guide for your case.   The tutorial you found is for local/IDE; the production-container equivalent is here: Trace agents deployed out...

  • 0 kudos
1 More Replies
cormierjohn
by New Contributor
  • 255 Views
  • 1 replies
  • 0 kudos

FMAPI Anthropic endpoint rejects requests with trailing assistant message — known limitation?

Hey all — looking for confirmation on a behavior I'm hitting on the Foundation Model API (pay-per-token) Anthropic-compatible endpoint, in case anyone else has worked around it.What I'm doing: serving Claude models through /serving-endpoints/anthropi...

  • 255 Views
  • 1 replies
  • 0 kudos
Latest Reply
stbjelcevic
Databricks Employee
  • 0 kudos

Hi @cormierjohn , Your understanding is correct. The validation rejecting a trailing assistant turn is happening at the FMAPI proxy layer before the request reaches Claude, so any client that uses Anthropic's prefill primitive will 400 against this e...

  • 0 kudos
skarangi_349
by New Contributor
  • 501 Views
  • 1 replies
  • 0 kudos

ai-dev-kit -Best practices and Scoped PATs

SpoilerHi,I am experimenting with ai-dev-kit and would like to ask whether there are any best practices, demos, or safety guardrails that were followed when using it. Specifically recommended settings for a limited-access PAT for dev-ai-toolkit.Thank...

  • 501 Views
  • 1 replies
  • 0 kudos
Latest Reply
Sumit_7
Esteemed Contributor
  • 0 kudos

@skarangi_349 As this is not a official DB product hence there's no such guide or docs. Best would be to follow the creators - https://www.youtube.com/@DustinVannoyYou may check other videos over youtube as well.

  • 0 kudos
frankc
by New Contributor III
  • 8322 Views
  • 7 replies
  • 6 kudos

OpenAi gpt-oss support for responses API?

I was looking forward to using there openAI responses API when the new gpt-oss models were added to Databricks.The openAI website states these model are compatible with the responses API.However, it does not seem to support it on Databricks. See erro...

  • 8322 Views
  • 7 replies
  • 6 kudos
Latest Reply
nishkambatta
New Contributor II
  • 6 kudos

Great discussion here—this highlights a real gap between the Chat Completions API and the newer Responses API. It’s clear the Responses API is the direction forward, especially with better multi-modal and tool-calling capabilities, but the lack of su...

  • 6 kudos
6 More Replies
juliechen
by New Contributor II
  • 200 Views
  • 1 replies
  • 1 kudos

Poll: With the current maturity of LLMs and Genie, how are you positioning Genie in your org

Do you see Genie (or similar AI analytics tools) as:1️⃣ Primarily for power users / data-savvy users to enhance their daily work by correlating multiple business signals to generate deeper insights and answer more complex questions for the organizati...

  • 200 Views
  • 1 replies
  • 1 kudos
Latest Reply
michael365
New Contributor III
  • 1 kudos

According to my experience both target groups can be adressed:Genie (spaces) are made for use by power usersEnabling Genie spaces via Databricks One* can be used by a broad organizationGenie Code is for technical persons e.g. data engineers, data sci...

  • 1 kudos
smith2
by New Contributor
  • 190 Views
  • 1 replies
  • 0 kudos

Need Help with This Issue in Databricks Workflow

Hi everyone,I’m currently working with Databricks and facing an issue related to my setup/workflow. I’ve tried checking the basic configurations, but I’m still not able to resolve it.Could someone please guide me on the best approach or point out wha...

  • 190 Views
  • 1 replies
  • 0 kudos
Latest Reply
Sumit_7
Esteemed Contributor
  • 0 kudos

@smith2 Please specify the issue faced and screenshots of errors will be helpful.

  • 0 kudos
martkev
by New Contributor III
  • 3465 Views
  • 1 replies
  • 0 kudos

Resolved! Error: "Invalid model name" in Databricks AI Gateway when setting up Vertex AI endpoint

Hi everyone,I'm trying to set up a new serving endpoint in Databricks using Google Cloud Vertex AI as the provider. I want to route it to claude-opus-4-6.However, as soon as I try to create it, I get the following UI error (see screenshot):"Invalid m...

martkev_0-1775828309497.png
  • 3465 Views
  • 1 replies
  • 0 kudos
Latest Reply
anuj_lathi
Databricks Employee
  • 0 kudos

Hi @martkev -- good news: your model name is correct. The Vertex AI model ID for Claude Opus 4.6 is indeed claude-opus-4-6 (per Anthropic's official documentation). The issue is on the Databricks side -- the UI enforces an internal allowlist of recog...

  • 0 kudos
trickywhitecat
by New Contributor
  • 566 Views
  • 1 replies
  • 0 kudos

Petition to update your documentation about inferencing LLMs and standardize LLM response format

TL;DRInconsistent & Undocumented Response Format: Contrary to Databricks documentation, enabling thinking.type for Gemini 2.5 Flash changes the ChatCompletion content field from a standard string to a list of dictionaries, breaking expected behavior....

  • 566 Views
  • 1 replies
  • 0 kudos
Latest Reply
Louis_Frolio
Databricks Employee
  • 0 kudos

Hey @trickywhitecat , You're right that when thinking.type is enabled for Gemini 2.5 Flash on a Databricks serving endpoint, the content field comes back as a list of dictionaries instead of a plain string. That breaks the expected OpenAI ChatComplet...

  • 0 kudos
uhuru_furuta
by Databricks Partner
  • 1389 Views
  • 4 replies
  • 0 kudos

Resolved! Genie integration in Dashboards fails with "GenericChatCompletionServiceException" (Free Edition)

I am experiencing a persistent error when using Genie Space integration within a Databricks Dashboard (v3/Lakeview). While Genie works perfectly as a standalone Space, it consistently fails when invoked via the Dashboard sidebar.Environment:Edition: ...

  • 1389 Views
  • 4 replies
  • 0 kudos
Latest Reply
emma_s
Databricks Employee
  • 0 kudos

Yes 100%, any user would need to have select permissions on the table that the Genie Space is created on top of.

  • 0 kudos
3 More Replies
souravg
by New Contributor
  • 830 Views
  • 2 replies
  • 1 kudos

Resolved! Feedback not showing up in Genie from Copilot Studio Genie Agent

Hi,We've created an Agent using Copilot Studio for Genie and integrated with Teams Channel.The feedback there is working and we can see the reactions in the Copilot Studio Analytics.But the feedback is not going to the actual genie space, neither the...

  • 830 Views
  • 2 replies
  • 1 kudos
Latest Reply
Ashwin_DSA
Databricks Employee
  • 1 kudos

Hi @souravg, @Ale_Armillotta is right. At the moment, Genie only records feedback (thumbs up/down, "Fix it", comments) when it’s given directly in the Genie UI. The public Genie Conversation APIs that Copilot Studio/Teams use don’t expose any endpoin...

  • 1 kudos
1 More Replies
rawalsparsh99
by New Contributor
  • 834 Views
  • 3 replies
  • 0 kudos

Knowledge Assistant is disabled

These features are disabled i have completed all the prerequisites for this.I have the admin rights and have enabled all the required preview both at the account and the workspace level.also I have set serverless budget quota for the workspace as wel...

Screenshot 2026-01-28 180951.png
  • 834 Views
  • 3 replies
  • 0 kudos
Latest Reply
SteveOstrowski
Databricks Employee
  • 0 kudos

Hi @rawalsparsh99, Knowledge Assistant is part of the Agent Bricks feature set, and there are a few specific requirements that all need to be in place simultaneously for the feature tiles to become active. Since you mentioned you have admin rights, p...

  • 0 kudos
2 More Replies
Ale_Armillotta
by Valued Contributor II
  • 828 Views
  • 1 replies
  • 1 kudos

Resolved! Voice Interface for Genie via Azure AI Foundry – 400 Bad Request with Cross-Subscription Setup

Hi everyone,I'm building a voice-enabled solution on Azure AI Foundry that allows users to interact with Genie using speech instead of text. The architecture involves exposing a Genie Space as a tool within Foundry, which works perfectly in my own en...

  • 828 Views
  • 1 replies
  • 1 kudos
Latest Reply
Ale_Armillotta
Valued Contributor II
  • 1 kudos

I found the issues.the resources were deployed into different tenant and subscriptions. Seems now it's possible only to autentificate with PAT and not with other auth. So for now I can use only my account and not a Service Principal

  • 1 kudos
Suheb
by Contributor
  • 611 Views
  • 1 replies
  • 0 kudos

Resolved! How do I get started with Databricks as a complete beginner?

How can a beginner start learning and using Databricks step by step?

  • 611 Views
  • 1 replies
  • 0 kudos
Latest Reply
joelramirezai
Databricks Employee
  • 0 kudos

You can use Databricks free edition to test Databricks and practice, even though the path will also depend in your goals as a professional. You can choose one or multiple of the following paths as guidance:  Platform, Data Engineering, Traditional ML...

  • 0 kudos