diff --git a/base/base/contest.py b/base/base/contest.py
index 40080ba0..9979ba34 100644
--- a/base/base/contest.py
+++ b/base/base/contest.py
@@ -98,11 +98,11 @@ def normalize(baseline_value: float, benchmark_value: float, metric_type: Metric
             return (relative_improvement * self.metric_weights.get(metric_type, 0)) / total_weight
 
         score = sum([
-            normalize(baseline.generation_time, benchmark.model.generation_time, MetricType.GENERATION_TIME),
-            normalize(baseline.size, benchmark.model.size, MetricType.SIZE),
-            normalize(baseline.vram_used, benchmark.model.vram_used, MetricType.VRAM_USED),
-            normalize(baseline.watts_used, benchmark.model.watts_used, MetricType.WATTS_USED),
-            normalize(baseline.load_time, benchmark.model.load_time, MetricType.LOAD_TIME)
+            normalize(baseline.generation_time, benchmark.metrics.generation_time, MetricType.GENERATION_TIME),
+            normalize(baseline.size, benchmark.metrics.size, MetricType.SIZE),
+            normalize(baseline.vram_used, benchmark.metrics.vram_used, MetricType.VRAM_USED),
+            normalize(baseline.watts_used, benchmark.metrics.watts_used, MetricType.WATTS_USED),
+            normalize(baseline.load_time, benchmark.metrics.load_time, MetricType.LOAD_TIME)
         ])
 
         return score * similarity * self.metric_weights.get(MetricType.SIMILARITY_SCORE, 0) / total_weight
diff --git a/base/testing/benchmarker.py b/base/testing/benchmarker.py
index 82a6d5f3..11520c57 100644
--- a/base/testing/benchmarker.py
+++ b/base/testing/benchmarker.py
@@ -124,7 +124,7 @@ def benchmark_submissions(self, contest: Contest, submissions: dict[Key, Reposit
             average_benchmarking_time = self.get_average_benchmarking_time()
             if average_benchmarking_time:
                 eta = (len(submissions) - len(self.benchmarks)) * average_benchmarking_time
-                logger.info(f"Average benchmark time: {average_benchmarking_time}, ETA: {timedelta(seconds=eta)}")
+                logger.info(f"Average benchmark time: {average_benchmarking_time:.2f}s, ETA: {timedelta(seconds=eta)}")
 
         if self._is_done(submissions):
             logger.info("Benchmarking complete")
diff --git a/validator/weight_setting/validator.py b/validator/weight_setting/validator.py
index f4515866..d636c0f9 100644
--- a/validator/weight_setting/validator.py
+++ b/validator/weight_setting/validator.py
@@ -185,7 +185,7 @@ def update_benchmarks(self, benchmarking_results: list[BenchmarkingResults]):
 
         if average_benchmarking_time:
             eta = (len(self.contest_state.submissions) - len(self.contest_state.benchmarks)) * average_benchmarking_time
-            logger.info(f"Average benchmark time: {average_benchmarking_time}, ETA: {timedelta(seconds=eta)}")
+            logger.info(f"Average benchmark time: {average_benchmarking_time:.2f}s, ETA: {timedelta(seconds=eta)}")
 
     def step(self):
         return self.contest_state.step if self.contest_state else 0