fix: support Azure provider with reasoning models

- Add azure, openrouter, bedrock, and ollama to VALIDATED_PROVIDERS array
- Add Azure reasoning models (GPT-5, o1, o3, o3-mini, o4-mini) to supported-models.json
- Implement automatic API endpoint detection for Azure reasoning models
- Add dual endpoint support (chat/completions vs responses) in AzureProvider
- Add smart URL adjustment logic for different Azure configurations
- Maintain backward compatibility with existing Azure setups

Fixes #638

🤖 Generated with [Claude Code](https://claude.ai/code)

Co-authored-by: Ralph Khreish <Crunchyman-ralph@users.noreply.github.com>

scripts/modules/supported-models.json

@@ -997,6 +997,67 @@
 			"allowed_roles": ["main", "fallback"],
 			"max_tokens": 16384,
 			"supported": true
+		},
+		{
+			"id": "gpt-5",
+			"swe_score": 0.749,
+			"cost_per_1m_tokens": {
+				"input": 5.0,
+				"output": 20.0
+			},
+			"allowed_roles": ["main", "fallback"],
+			"max_tokens": 100000,
+			"temperature": 1,
+			"supported": true,
+			"api_type": "responses"
+		},
+		{
+			"id": "o1",
+			"swe_score": 0.489,
+			"cost_per_1m_tokens": {
+				"input": 15.0,
+				"output": 60.0
+			},
+			"allowed_roles": ["main"],
+			"max_tokens": 100000,
+			"supported": true,
+			"api_type": "responses"
+		},
+		{
+			"id": "o3",
+			"swe_score": 0.5,
+			"cost_per_1m_tokens": {
+				"input": 2.0,
+				"output": 8.0
+			},
+			"allowed_roles": ["main", "fallback"],
+			"max_tokens": 100000,
+			"supported": true,
+			"api_type": "responses"
+		},
+		{
+			"id": "o3-mini",
+			"swe_score": 0.493,
+			"cost_per_1m_tokens": {
+				"input": 1.1,
+				"output": 4.4
+			},
+			"allowed_roles": ["main"],
+			"max_tokens": 100000,
+			"supported": true,
+			"api_type": "responses"
+		},
+		{
+			"id": "o4-mini",
+			"swe_score": 0.45,
+			"cost_per_1m_tokens": {
+				"input": 1.1,
+				"output": 4.4
+			},
+			"allowed_roles": ["main", "fallback"],
+			"max_tokens": 100000,
+			"supported": true,
+			"api_type": "responses"
 		}
 	],
 	"bedrock": [

src/ai-providers/azure.js

@@ -5,6 +5,7 @@
 
 import { createAzure } from '@ai-sdk/azure';
 import { BaseAIProvider } from './base-provider.js';
+import MODEL_MAP from '../../scripts/modules/supported-models.json' with { type: 'json' };
 
 export class AzureProvider extends BaseAIProvider {
 	constructor() {
@@ -37,21 +38,65 @@ export class AzureProvider extends BaseAIProvider {
 		}
 	}
 
+	/**
+	 * Determines if a model requires the responses API endpoint instead of chat/completions
+	 * @param {string} modelId - The model ID to check
+	 * @returns {boolean} True if the model needs the responses API
+	 */
+	isReasoningModel(modelId) {
+		const azureModels = MODEL_MAP.azure || [];
+		const modelDef = azureModels.find(m => m.id === modelId);
+		return modelDef?.api_type === 'responses';
+	}
+
+	/**
+	 * Adjusts the base URL for reasoning models that need the responses endpoint
+	 * @param {string} baseURL - Original base URL
+	 * @param {string} modelId - Model ID
+	 * @returns {string} Adjusted base URL
+	 */
+	adjustBaseURL(baseURL, modelId) {
+		if (!this.isReasoningModel(modelId)) {
+			return baseURL;
+		}
+
+		// Convert chat/completions URL to responses URL for reasoning models
+		if (baseURL.includes('/chat/completions')) {
+			return baseURL.replace('/chat/completions', '/responses');
+		}
+
+		// If baseURL ends with deployments/<model-name>, add responses endpoint
+		if (baseURL.includes('/deployments/')) {
+			return baseURL.replace(/\/deployments\/[^\/]+$/, '/responses');
+		}
+
+		// If baseURL is just the base, ensure it ends with /responses
+		if (!baseURL.endsWith('/responses')) {
+			return baseURL.replace(/\/$/, '') + '/responses';
+		}
+
+		return baseURL;
+	}
+
 	/**
 	 * Creates and returns an Azure OpenAI client instance.
 	 * @param {object} params - Parameters for client initialization
 	 * @param {string} params.apiKey - Azure OpenAI API key
 	 * @param {string} params.baseURL - Azure OpenAI endpoint URL (from .taskmasterconfig global.azureBaseURL or models.[role].baseURL)
+	 * @param {string} params.modelId - Model ID (used to determine API endpoint)
 	 * @returns {Function} Azure OpenAI client function
 	 * @throws {Error} If required parameters are missing or initialization fails
 	 */
 	getClient(params) {
 		try {
-			const { apiKey, baseURL } = params;
+			const { apiKey, baseURL, modelId } = params;
+
+			// Adjust base URL for reasoning models
+			const adjustedBaseURL = this.adjustBaseURL(baseURL, modelId);
 
 			return createAzure({
 				apiKey,
-				baseURL
+				baseURL: adjustedBaseURL
 			});
 		} catch (error) {
 			this.handleError('client initialization', error);

src/constants/providers.js

@@ -11,7 +11,11 @@ export const VALIDATED_PROVIDERS = [
 	'perplexity',
 	'xai',
 	'groq',
-	'mistral'
+	'mistral',
+	'azure',
+	'openrouter',
+	'bedrock',
+	'ollama'
 ];
 
 // Custom providers object for easy named access