Consuming a custom Azure AI Foundry Endpoint #7166
scara1701
started this conversation in
Help Wanted
Replies: 0 comments
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
-
I managed to get our OpenAI Azure endpoint working with the 4o-mini model. Which works on librechat with following config:
azureOpenAI: groups: - group: "somename-gpt4o" apiKey: "somekey" serverless: true baseURL: "https://somename.openai.azure.com/openai/deployments/somename-gpt4o-mini" version: "2023-12-01-preview" models: gpt-4o-mini: deploymentName: somename-gpt4o-mini version: "2024-07-18"
I also have published an Azure AI Prompt flow (a chat flow) to an endpoint on azure.
Now I would like to utilize this endpoint on a libreChat interface. I am however a bit confused on how to implement this.
As an input the flow expects a 'Chat_history' (a list of previous messages) and 'question'.
custom: - name: "anothername" apiKey: "anotherkey" baseURL: "https://anothername.westeurope.inference.ml.azure.com/score" directEndpoint: true models: default: ["modelname"]
The endpoint responds with an error 424, due toe the missing input fields ['chat_history', 'question'] . How do I add these to my request in the model?
Am I supposed to use the addParams option? If so, how can I refer to the chat history and most recent question?
This is the swagger generated by Azure AI for the endpoint.
{ "components": { "securitySchemes": { "bearerAuth": { "scheme": "bearer", "type": "http" } } }, "info": { "title": "Promptflow[model] API", "version": "1.0.0", "x-chat-history": "chat_history", "x-chat-input": "question", "x-chat-output": "answer", "x-flow-name": "model", "x-flow-type": "chat" }, "openapi": "3.0.0", "paths": { "/feedback": { "parameters": [ { "description": "flatten the feedback data into traced data", "in": "query", "name": "flatten", "required": false, "schema": { "type": "boolean" } } ], "post": { "requestBody": { "content": { "application/json": { "schema": { "additionalProperties": {}, "type": "object" } } }, "description": "promptflow feedback data", "required": true }, "responses": { "200": { "content": { "application/json": { "schema": { "additionalProperties": {}, "type": "object" } } }, "description": "successful operation" }, "400": { "description": "Invalid input" }, "default": { "description": "unexpected error" } }, "summary": "collect promptflow feedback" } }, "/score": { "post": { "requestBody": { "content": { "application/json": { "example": {}, "schema": { "properties": { "chat_history": { "items": { "additionalProperties": {}, "type": "object" }, "type": "array" }, "question": { "type": "string" } }, "required": [ "chat_history", "question" ], "type": "object" } } }, "description": "promptflow input data", "required": true }, "responses": { "200": { "content": { "application/json": { "schema": { "properties": { "answer": { "type": "string" } }, "type": "object" } } }, "description": "successful operation" }, "400": { "description": "Invalid input" }, "default": { "description": "unexpected error" } }, "summary": "run promptflow: model with an given input" } } }, "security": [ { "bearerAuth": [] } ] }
Beta Was this translation helpful? Give feedback.
All reactions