Refactor Azure OpenAI Initialization and Model Handling #510
Conversation
romanlum
changed the title
|
This pull request has been marked as stale due to inactivity. Please update it to keep it open. |
|
would be cool when you could merge this PR because otherwise azure users are not able to use another model (e.g. gpt-4.1-nano) because of the hardcoded gpt-4o-mini |
| @@ -135,7 +135,7 @@ class SetupService { | |||
| deploymentName: deploymentName, | |||
| apiVersion: apiVersion }); | |||
| const response = await openai.chat.completions.create({ | |||
| model: "gpt-4o-mini", | |||
| model: deploymentName, //azure openai uses deployment name as model parameter | |||
There was a problem hiding this comment.
At this point your suggestion is correct.
| @@ -228,7 +228,7 @@ class AzureOpenAIService { | |||
|
|
|||
| // Make API request | |||
| const response = await this.client.chat.completions.create({ | |||
| model: process.env.AZURE_DEPLOYMENT_NAME, | |||
| model: config.azure.deploymentName, //azure openai uses deployment name as model parameter | |||
There was a problem hiding this comment.
It uses the deploymentName already here. Just loaded from ENV
There was a problem hiding this comment.
yes but to make it consistent i used the configuration value here
| @@ -129,7 +129,7 @@ class AzureOpenAIService { | |||
| await writePromptToFile(systemPrompt, truncatedContent); | |||
|
|
|||
| const response = await this.client.chat.completions.create({ | |||
| model: model, | |||
| model: config.azure.deploymentName, //azure openai uses deployment name as model parameter | |||
There was a problem hiding this comment.
Model = deploymentName here. See line 65
There was a problem hiding this comment.
yes but to make it consistent i used the configuration value here
There was a problem hiding this comment.
No worries. That was not meant in a bad way. You are absolutely right with your improvements.
It were only review comments for documentary reasons.
|
This pull request has been marked as stale due to inactivity. Please update it to keep it open. |
This PR refactors the Azure OpenAI integration and allows other models than gpt-4o-mini (e.g. gpt-4.1-nano).
Key improvements include:
Simplified client initialization: Removed redundant provider checks and streamlined logic to initialize the Azure OpenAI client only when config.aiProvider === 'azure'.
Corrected model usage: Replaced hardcoded model names with config.azure.deploymentName in all chat.completions.create calls, as Azure uses deployment names instead of model names at runtime.
Code cleanup: Removed unused or outdated initialization blocks for other providers (ollama, custom, etc.) within the Azure service context.