9 results found
-
Grounding
Enable VertexAI web access for googleSearchRetrieval() and Grounding to work on the FirebaseVertexAI.
7 votes -
Vertex AI only for authenticated users
Currently, App Check is the only mechanism to control access to the Vertex API. This is not always practical and does not allow per-user access control.
Please add a mechanism similar to Security Rules for Firestore, to enforce only authenticated Firebase users to access the AI APIs.6 votes -
1 vote
-
Enhanced Genkit Integration in Firebase Studio for Context-Aware AI Development
- Introduction & Problem Statement: Firebase Studio aims to be a premier cloud-based IDE for building modern applications, including those powered by AI using Genkit and Vertex AI. Currently, while Genkit provides powerful tools for defining AI flows (data indexing, RAG, code generation), the workflow for developing, testing, and leveraging these flows locally within Firebase Studio presents significant friction. Users seeking to build sophisticated, project-aware AI assistants (that go beyond the built-in chat features) face challenges: Discoverability of Execution Mechanisms: The intended method for locally running custom Genkit flows (defined with defineFlow) or launching the Genkit Developer UI within the Studio…
0 votes -
1 vote
-
Allow access to Gemini preview models
Currently the latest stable or other stable Gemini 1.5 models are available only (see https://firebase.google.com/docs/vertex-ai/gemini-models#available-model-names). These lag behind with features such as function calling ability (performance), structured output (according to the feature matrix Gemini 1.5 Flash does not support that but we know it has multiple versions already capable of). Why cannot we access the preview model we'd like?
4 votes -
Train A Model Through API
I don't know if this is already exisiting but i think it should be possible to train a custom model through the vertex AI API for firebase.
3 votes -
Add input and output tokens used and model to the response object.
One of the very useful things about the Open AI API is that the response includes the amount of input and output tokens used , as well as the model that was hit. I am recording this data per call in my firebase instance for my app along with time of processing to monitor performance. I also add the appropriate cost per model in there and so I dynamically calculate the costs per model call as well. This helps me make business case decisions on how I use the AI.
It would be great to have that available in the Vertex AI API…2 votes -
Automated Function Calling for Kotlin
Allow the SDK to call automatically the functions that the model requires before to proceed with the prompt
2 votes
- Don't see your idea?