diff --git a/scripts/mma_exec.py b/scripts/mma_exec.py
index 61c4ba9..2fedb66 100644
--- a/scripts/mma_exec.py
+++ b/scripts/mma_exec.py
@@ -2,26 +2,36 @@ import argparse
 import subprocess
 import json
 
+def get_model_for_role(role: str) -> str:
+    """Returns the specific model to use for a given tier role."""
+    if role == 'tier1-orchestrator' or role == 'tier1':
+        return 'gemini-3.1-pro-preview'
+    elif role == 'tier2-tech-lead' or role == 'tier2':
+        return 'gemini-3.0-flash-preview'
+    else:
+        return 'gemini-2.5-flash-lite'
+
 def get_role_documents(role: str) -> list[str]:
-    if role == 'tier1':
+    if role == 'tier1-orchestrator' or role == 'tier1':
         return ['conductor/product.md', 'conductor/product-guidelines.md']
-    elif role == 'tier2':
+    elif role == 'tier2-tech-lead' or role == 'tier2':
         return ['conductor/tech-stack.md', 'conductor/workflow.md']
-    elif role == 'tier3':
+    elif role == 'tier3-worker' or role == 'tier3':
         return ['conductor/workflow.md']
     return []
 
 def execute_agent(role: str, prompt: str, docs: list[str]) -> str:
+    model = get_model_for_role(role)
     command_text = f"Activate the mma-{role} skill. {prompt}"
     for doc in docs:
         command_text += f" @{doc}"
     
-    cmd = ['gemini', '-p', command_text, '--output-format', 'json', '--model', 'gemini-2.5-flash-lite']
+    cmd = ['gemini', '-p', command_text, '--output-format', 'json', '--model', model]
     try:
         process = subprocess.run(cmd, capture_output=True, text=True, shell=True)
         if not process.stdout and process.stderr:
             return f"Error: {process.stderr}"
-            
+
         stdout = process.stdout
         start_index = stdout.find('{')
         if start_index != -1:
@@ -39,7 +49,7 @@ def create_parser():
     parser = argparse.ArgumentParser(description="MMA Execution Script")
     parser.add_argument(
         "--role", 
-        choices=['tier1', 'tier2', 'tier3', 'tier4'], 
+        choices=['tier1', 'tier2', 'tier3', 'tier4', 'tier1-orchestrator', 'tier2-tech-lead', 'tier3-worker', 'tier4-qa'], 
         required=True, 
         help="The tier role to execute"
     )
diff --git a/tests/test_mma_exec.py b/tests/test_mma_exec.py
index 56964e6..f20f024 100644
--- a/tests/test_mma_exec.py
+++ b/tests/test_mma_exec.py
@@ -1,6 +1,6 @@
 import pytest
 from unittest.mock import patch, MagicMock
-from scripts.mma_exec import create_parser, get_role_documents, execute_agent
+from scripts.mma_exec import create_parser, get_role_documents, execute_agent, get_model_for_role
 
 def test_parser_role_choices():
     """Test that the parser accepts valid roles and the prompt argument."""
@@ -39,16 +39,24 @@ def test_get_role_documents():
     assert get_role_documents('tier3') == ['conductor/workflow.md']
     assert get_role_documents('tier4') == []
 
+def test_get_model_for_role():
+    """Test that get_model_for_role returns the correct model for each role."""
+    assert get_model_for_role('tier1-orchestrator') == 'gemini-3.1-pro-preview'
+    assert get_model_for_role('tier2-tech-lead') == 'gemini-3.0-flash-preview'
+    assert get_model_for_role('tier3-worker') == 'gemini-2.5-flash-lite'
+    assert get_model_for_role('tier4-qa') == 'gemini-2.5-flash-lite'
+
 def test_execute_agent():
     """
     Test that execute_agent calls subprocess.run with the correct gemini CLI arguments
-    for context amnesia.
+    including the model specified for the role.
     """
-    role = "tier3"
+    role = "tier3-worker"
     prompt = "Write a unit test."
     docs = ["file1.py", "docs/spec.md"]
     
-    expected_gemini_arg = "Activate the mma-tier3 skill. Write a unit test. @file1.py @docs/spec.md"
+    expected_gemini_arg = "Activate the mma-tier3-worker skill. Write a unit test. @file1.py @docs/spec.md"
+    expected_model = "gemini-2.5-flash-lite"
     mock_stdout = "Mocked AI Response"
 
     with patch("subprocess.run") as mock_run:
@@ -66,7 +74,13 @@ def test_execute_agent():
         assert cmd_list[0] == "gemini"
         assert cmd_list[1] == "-p"
         assert cmd_list[2] == expected_gemini_arg
+
+        # Verify the correct model is passed via --model flag
+        assert "--model" in cmd_list
+        model_idx = cmd_list.index("--model")
+        assert cmd_list[model_idx + 1] == expected_model
+
         assert kwargs.get("capture_output") is True
         assert kwargs.get("text") is True
-        
+
         assert result == mock_stdout
\ No newline at end of file