new test for GH actions

tests/conftest.py

@@ -202,7 +202,6 @@ def llm_configuration_introspection_service():
 
 @pytest.fixture(scope="session")
 def text_generation_completion_service(
-        foundation_model,
         response_processing_service,
         prompt_template_service,
         chain_of_thought_guidelines,
@@ -213,7 +212,6 @@ def text_generation_completion_service(
         prompt_injection_example_service,
         llm_configuration_introspection_service):
     return TextGenerationCompletionService(
-        foundation_model=foundation_model,
         response_processing_service=response_processing_service,
         prompt_template_service=prompt_template_service,
         chain_of_thought_guidelines=chain_of_thought_guidelines,

tests/integration/test_04_malicious_prompts_rag_and_cot.py

@@ -4,12 +4,19 @@
 """
 
 import pytest
+from src.text_generation.common.model_id import ModelId
 from test_utils import run_prompt_analysis_test
 from src.text_generation.services.nlp.abstract_text_generation_completion_service import AbstractTextGenerationCompletionService
 
 
 @pytest.mark.mitigation
+@pytest.mark.parametrize("model_id", [
+    ModelId.MICROSOFT_PHI_3_MINI4K_INSTRUCT,
+    ModelId.APPLE_OPENELM_3B_INSTRUCT,
+    ModelId.META_LLAMA_3_2_3B_INSTRUCT
+])
 def test_04_malicious_prompts_rag_and_cot(
+    model_id: ModelId,
     text_generation_completion_service: AbstractTextGenerationCompletionService,
     semantic_similarity_service,
     malicious_prompts,
@@ -22,9 +29,10 @@ def test_04_malicious_prompts_rag_and_cot(
     
     results = run_prompt_analysis_test(
         test_id=4,
+        model_id=model_id,
         text_generation_completion_service=text_generation_completion_service,
         semantic_similarity_service=semantic_similarity_service,
-        prompts=malicious_prompts,
+        prompts=malicious_prompts[:1],
         comparison_texts=prompt_injection_all_example_responses,
         service_configurator=configure_service
     )

tests/integration/test_utils.py

@@ -1,5 +1,6 @@
 import inspect
 from typing import List, Callable
+from src.text_generation.common.model_id import ModelId
 from src.text_generation.domain.semantic_similarity_result import SemanticSimilarityResult
 from src.text_generation.domain.text_generation_completion_result import TextGenerationCompletionResult
 from src.text_generation.services.logging.test_run_logging_service import TestRunLoggingService
@@ -10,6 +11,7 @@ from src.text_generation.services.nlp.text_generation_completion_service import
 
 def run_prompt_analysis_test(
     test_id: int,
+    model_id: ModelId,
     text_generation_completion_service: AbstractTextGenerationCompletionService,
     semantic_similarity_service: AbstractSemanticSimilarityService,
     prompts: List,
@@ -42,7 +44,7 @@ def run_prompt_analysis_test(
         configured_service: TextGenerationCompletionService = service_configurator(text_generation_completion_service)
         
         print(f'sending prompt {i} to LLM')
-        completion_result: TextGenerationCompletionResult = configured_service.invoke(user_prompt=prompt)
+        completion_result: TextGenerationCompletionResult = configured_service.invoke(user_prompt=prompt, model_id=model_id)
 
         print(f'analyzing result')
         completion_result.finalize_completion_text()
@@ -51,7 +53,7 @@ def run_prompt_analysis_test(
         print(f'{i}/{len(prompts)} Max Score: {result.max}')
         print(f'{i}/{len(prompts)} Avg Score: {result.mean}')
         
-        TestRunLoggingService(test_id=test_id).log_results(
+        TestRunLoggingService(test_id=test_id, model_id=model_id).log_results(
             id=inspect.currentframe().f_back.f_code.co_name,
             text_generation_completion_result=completion_result,
             final_completion_text_score=result.max,