docs: Add comments and expand test coverage

- Add documentation comments for MODEL env var
- Expand test coverage for Azure MODEL env var
- Add test for missing MODEL env var case

Co-Authored-By: Joe Moura <joao@crewai.com>

src/crewai/cli/constants.py

@@ -66,7 +66,7 @@ ENV_VARS = {
     "azure": [
         {
             "prompt": "Enter your Azure deployment name (must start with 'azure/')",
-            "key_name": "MODEL",
+            "key_name": "MODEL",  # Uppercase MODEL used for consistency across environment variables
         },
         {
             "prompt": "Enter your AZURE API key (press Enter to skip)",
@@ -84,7 +84,7 @@ ENV_VARS = {
     "cerebras": [
         {
             "prompt": "Enter your Cerebras model name (must start with 'cerebras/')",
-            "key_name": "MODEL",
+            "key_name": "MODEL",  # Uppercase MODEL used for consistency across environment variables
         },
         {
             "prompt": "Enter your Cerebras API version (press Enter to skip)",

tests/agent_test.py

@@ -22,19 +22,33 @@ from crewai.utilities.events import Emitter
 
 
 def test_agent_azure_model_env_var():
+    """Test Azure MODEL environment variable handling with various cases."""
     # Store original environment variables
     original_model = os.environ.get("MODEL")
 
-    # Test with uppercase MODEL
-    os.environ["MODEL"] = "azure/test-model"
+    test_cases = [
+        ("azure/test-model", "azure/test-model"),  # Valid case
+        ("azure/minimal", "azure/minimal"),  # Another valid case
+    ]
+
+    for input_model, expected_model in test_cases:
+        # Set test MODEL value
+        os.environ["MODEL"] = input_model
+        agent = Agent(role="test role", goal="test goal", backstory="test backstory")
+        assert agent.llm.model == expected_model
+
+    # Test missing MODEL env var
+    if "MODEL" in os.environ:
+        del os.environ["MODEL"]
     agent = Agent(role="test role", goal="test goal", backstory="test backstory")
-    assert agent.llm.model == "azure/test-model"
+    assert agent.llm.model == "gpt-4o-mini"  # Default model
 
     # Clean up environment variables
     if original_model:
         os.environ["MODEL"] = original_model
     else:
-        del os.environ["MODEL"]
+        if "MODEL" in os.environ:
+            del os.environ["MODEL"]
 
 
 def test_agent_llm_creation_with_env_vars():