diff --git a/lib/cli/src/crewai_cli/benchmark.py b/lib/cli/src/crewai_cli/benchmark.py
index fc3d9ebc8..0128d9a49 100644
--- a/lib/cli/src/crewai_cli/benchmark.py
+++ b/lib/cli/src/crewai_cli/benchmark.py
@@ -3,13 +3,14 @@
 from __future__ import annotations
 
 import asyncio
+from collections.abc import Callable
 import json
+from pathlib import Path
 import re
 import time
-from pathlib import Path
-from typing import Any, Callable
+from typing import Any
 
-from pydantic import BaseModel, Field
+from pydantic import BaseModel
 
 
 class BenchmarkCase(BaseModel):
@@ -340,14 +341,16 @@ async def run_benchmark(
     return dict(zip(models, all_results))
 
 
-class suppress_benchmark_output:
+class SuppressBenchmarkOutput:
     """Context manager that silences console formatter and noisy logging during benchmarks."""
 
     def __enter__(self):
         import logging
         self._saved_formatter = None
         try:
-            from crewai.events.listeners.tracing.trace_listener import TraceCollectionListener
+            from crewai.events.listeners.tracing.trace_listener import (
+                TraceCollectionListener,
+            )
             listener = TraceCollectionListener._instance
             if listener:
                 self._saved_formatter = listener.formatter
@@ -366,7 +369,9 @@ class suppress_benchmark_output:
             lg.setLevel(level)
         if self._saved_formatter is not None:
             try:
-                from crewai.events.listeners.tracing.trace_listener import TraceCollectionListener
+                from crewai.events.listeners.tracing.trace_listener import (
+                    TraceCollectionListener,
+                )
                 listener = TraceCollectionListener._instance
                 if listener:
                     listener.formatter = self._saved_formatter
@@ -374,7 +379,7 @@ class suppress_benchmark_output:
                 pass
 
 
-class artifacts_sandbox:
+class ArtifactsSandbox:
     """Context manager that chdirs into tests/artifacts/ for the benchmark run.
 
     Any files created by agent tools land in the artifacts directory instead of
diff --git a/lib/cli/src/crewai_cli/cli.py b/lib/cli/src/crewai_cli/cli.py
index d317b2909..4b2eb557a 100644
--- a/lib/cli/src/crewai_cli/cli.py
+++ b/lib/cli/src/crewai_cli/cli.py
@@ -784,11 +784,11 @@ def _test_new_agents(
         fg="cyan", bold=True,
     )
 
-    from crewai_cli.benchmark import artifacts_sandbox, suppress_benchmark_output
+    from crewai_cli.benchmark import ArtifactsSandbox, SuppressBenchmarkOutput
 
     progress.start()
     try:
-        with artifacts_sandbox(), suppress_benchmark_output():
+        with ArtifactsSandbox(), SuppressBenchmarkOutput():
             all_results = asyncio.run(_run_all())
     finally:
         progress.stop()
@@ -803,7 +803,7 @@ def _test_new_agents(
             continue
 
         agents_tested += 1
-        for model_name, results in result.items():
+        for results in result.values():
             failed = [r for r in results if r.score < job["threshold"]]
             if failed:
                 all_passed = False
@@ -1612,12 +1612,12 @@ def benchmark(
     click.echo(f"Judge model: {judge_model}")
     click.echo()
 
-    from crewai_cli.benchmark import artifacts_sandbox, suppress_benchmark_output
+    from crewai_cli.benchmark import ArtifactsSandbox, SuppressBenchmarkOutput
 
     progress = _BenchmarkLiveProgress(console=_con)
     progress.start()
     try:
-        with artifacts_sandbox(), suppress_benchmark_output():
+        with ArtifactsSandbox(), SuppressBenchmarkOutput():
             results_by_model = asyncio.run(
                 run_benchmark(
                     agent_def=agent_path,