diff --git a/dingo/model/llm/llm_keyword_matcher.py b/dingo/model/llm/llm_keyword_matcher.py
index 440e58fb..d3948c12 100644
--- a/dingo/model/llm/llm_keyword_matcher.py
+++ b/dingo/model/llm/llm_keyword_matcher.py
@@ -9,19 +9,12 @@
 from typing import List
 
 from dingo.io import Data
+from dingo.io.output.eval_detail import EvalDetail, QualityLabel
 from dingo.model import Model
 from dingo.model.llm.base_openai import BaseOpenAI
 from dingo.utils import log
 from dingo.utils.exception import ConvertJsonError
 
-# Import EvalDetail for dev branch compatibility, fallback to ModelRes for main branch
-try:
-    from dingo.io.output.eval_detail import EvalDetail, QualityLabel
-    USE_EVAL_DETAIL = True
-except ImportError:
-    from dingo.model.modelres import ModelRes
-    USE_EVAL_DETAIL = False
-
 # Complete synonym mapping for keyword normalization
 SYNONYM_MAP = {
     "k8s": "Kubernetes",
@@ -185,7 +178,7 @@ def _build_prompt(jd_text: str, resume_text: str) -> str:
 
     @classmethod
     def process_response(cls, response: str):
-        """Process LLM response. Returns EvalDetail (dev) or ModelRes (main)."""
+        """Process LLM response. Returns EvalDetail."""
         log.info(f"Raw LLM response: {response}")
 
         # Extract think content and clean response
@@ -213,29 +206,16 @@ def process_response(cls, response: str):
 
         log.info(f"Keyword match score: {score:.1%}, threshold: {cls.threshold:.0%}")
 
-        # Return appropriate result type based on branch
-        if USE_EVAL_DETAIL:
-            result = EvalDetail(metric=cls.__name__)
-            result.score = score
-            result.reason = [reason]
-            if score >= cls.threshold:
-                result.status = False
-                result.label = [QualityLabel.QUALITY_GOOD]
-            else:
-                result.status = True
-                result.label = [f"QUALITY_BAD.{cls.__name__}"]
+        # Return EvalDetail result
+        result = EvalDetail(metric=cls.__name__)
+        result.score = score
+        result.reason = [reason]
+        if score >= cls.threshold:
+            result.status = False
+            result.label = [QualityLabel.QUALITY_GOOD]
         else:
-            result = ModelRes()
-            result.score = score
-            result.reason = [reason]
-            if score >= cls.threshold:
-                result.error_status = False
-                result.type = "KEYWORD_MATCH_GOOD"
-                result.name = "MATCH_GOOD"
-            else:
-                result.error_status = True
-                result.type = "KEYWORD_MATCH_LOW"
-                result.name = "MATCH_LOW"
+            result.status = True
+            result.label = [f"QUALITY_BAD.{cls.__name__}"]
 
         return result
 
@@ -346,38 +326,22 @@ def _generate_reason(cls, jd_analysis: dict, keyword_analysis: List[dict], score
 
     @classmethod
     def eval(cls, input_data: Data):
-        """Override eval to validate inputs. Returns EvalDetail (dev) or ModelRes (main)."""
+        """Override eval to validate inputs. Returns EvalDetail."""
         # Validate that content (resume) is provided
         if not input_data.content:
-            if USE_EVAL_DETAIL:
-                result = EvalDetail(metric=cls.__name__)
-                result.status = True
-                result.label = [f"QUALITY_BAD.{cls.__name__}"]
-                result.reason = ["Resume text (content) is required but was not provided"]
-                return result
-            else:
-                return ModelRes(
-                    error_status=True,
-                    type="KEYWORD_MATCH_ERROR",
-                    name="MISSING_RESUME",
-                    reason=["Resume text (content) is required but was not provided"]
-                )
+            result = EvalDetail(metric=cls.__name__)
+            result.status = True
+            result.label = [f"QUALITY_BAD.{cls.__name__}"]
+            result.reason = ["Resume text (content) is required but was not provided"]
+            return result
 
         # Validate that prompt (JD) is provided
         if not input_data.prompt:
-            if USE_EVAL_DETAIL:
-                result = EvalDetail(metric=cls.__name__)
-                result.status = True
-                result.label = [f"QUALITY_BAD.{cls.__name__}"]
-                result.reason = ["Job description (prompt) is required but was not provided"]
-                return result
-            else:
-                return ModelRes(
-                    error_status=True,
-                    type="KEYWORD_MATCH_ERROR",
-                    name="MISSING_JD",
-                    reason=["Job description (prompt) is required but was not provided"]
-                )
+            result = EvalDetail(metric=cls.__name__)
+            result.status = True
+            result.label = [f"QUALITY_BAD.{cls.__name__}"]
+            result.reason = ["Job description (prompt) is required but was not provided"]
+            return result
 
         # Call parent eval method
         return super().eval(input_data)
diff --git a/dingo/model/llm/llm_resume_optimizer.py b/dingo/model/llm/llm_resume_optimizer.py
index 5bc243f8..85794ee4 100644
--- a/dingo/model/llm/llm_resume_optimizer.py
+++ b/dingo/model/llm/llm_resume_optimizer.py
@@ -9,19 +9,12 @@
 from typing import List, Tuple
 
 from dingo.io import Data
+from dingo.io.output.eval_detail import EvalDetail, QualityLabel
 from dingo.model import Model
 from dingo.model.llm.base_openai import BaseOpenAI
 from dingo.utils import log
 from dingo.utils.exception import ConvertJsonError
 
-# Import EvalDetail for dev branch compatibility, fallback to ModelRes for main branch
-try:
-    from dingo.io.output.eval_detail import EvalDetail, QualityLabel
-    USE_EVAL_DETAIL = True
-except ImportError:
-    from dingo.model.modelres import ModelRes
-    USE_EVAL_DETAIL = False
-
 
 @Model.llm_register("LLMResumeOptimizer")
 class LLMResumeOptimizer(BaseOpenAI):
@@ -191,7 +184,7 @@ def _parse_match_report(cls, match_report) -> Tuple[List[str], List[str], List[s
 
     @classmethod
     def process_response(cls, response: str):
-        """Process LLM response. Returns EvalDetail (dev) or ModelRes (main)."""
+        """Process LLM response. Returns EvalDetail."""
         log.info(f"Raw LLM response length: {len(response)} chars")
 
         # Clean response
@@ -210,22 +203,13 @@ def process_response(cls, response: str):
         # Generate reason text
         reason = cls._generate_reason(optimization_summary, section_changes, overall_improvement)
 
-        # Return appropriate result type based on branch
-        if USE_EVAL_DETAIL:
-            result = EvalDetail(metric=cls.__name__)
-            result.status = False
-            result.label = [QualityLabel.QUALITY_GOOD]
-            result.reason = [reason]
-            # Store full response for downstream use (using extra field)
-            result.optimized_content = response_json
-        else:
-            result = ModelRes()
-            result.error_status = False
-            result.type = "RESUME_OPTIMIZED"
-            result.name = "OPTIMIZATION_COMPLETE"
-            result.reason = [reason]
-            # Store full response for downstream use
-            result.optimized_content = response_json
+        # Return EvalDetail result
+        result = EvalDetail(metric=cls.__name__)
+        result.status = False
+        result.label = [QualityLabel.QUALITY_GOOD]
+        result.reason = [reason]
+        # Store full response for downstream use (using extra field)
+        result.optimized_content = response_json
 
         return result
 
@@ -287,22 +271,14 @@ def _generate_reason(cls, summary: dict, changes: List[dict], overall: str) -> s
 
     @classmethod
     def eval(cls, input_data: Data):
-        """Override eval to validate inputs. Returns EvalDetail (dev) or ModelRes (main)."""
+        """Override eval to validate inputs. Returns EvalDetail."""
         # Validate that content (resume) is provided
         if not input_data.content:
-            if USE_EVAL_DETAIL:
-                result = EvalDetail(metric=cls.__name__)
-                result.status = True
-                result.label = [f"QUALITY_BAD.{cls.__name__}"]
-                result.reason = ["Resume text (content) is required but was not provided"]
-                return result
-            else:
-                return ModelRes(
-                    error_status=True,
-                    type="RESUME_OPTIMIZER_ERROR",
-                    name="MISSING_RESUME",
-                    reason=["Resume text (content) is required but was not provided"]
-                )
+            result = EvalDetail(metric=cls.__name__)
+            result.status = True
+            result.label = [f"QUALITY_BAD.{cls.__name__}"]
+            result.reason = ["Resume text (content) is required but was not provided"]
+            return result
 
         # Call parent eval method
         return super().eval(input_data)