diff --git a/src/guidellm/benchmark/output.py b/src/guidellm/benchmark/output.py
index 95b51d70..802f2758 100644
--- a/src/guidellm/benchmark/output.py
+++ b/src/guidellm/benchmark/output.py
@@ -579,6 +579,13 @@ async def finalize(self, report: GenerativeBenchmarksReport) -> Path:
                 benchmark_headers: list[str] = []
                 benchmark_values: list[str | float | list[float]] = []
 
+                # Add basic run description info
+                desc_headers, desc_values = (
+                    self._get_benchmark_desc_headers_and_values(benchmark)
+                )
+                benchmark_headers.extend(desc_headers)
+                benchmark_values.extend(desc_values)
+
                 # Add status-based metrics
                 for status in StatusDistributionSummary.model_fields:
                     status_headers, status_values = (
@@ -684,6 +691,21 @@ def _get_benchmark_status_metrics_stats(
         ]
         return headers, values
 
+    def _get_benchmark_extras_headers_and_values(
+        self, benchmark: GenerativeBenchmark,
+    ) -> tuple[list[str], list[str]]:
+        headers = ["Profile", "Backend", "Generator Data"]
+        values: list[str] = [
+            benchmark.benchmarker.profile.model_dump_json(),
+            json.dumps(benchmark.benchmarker.backend),
+            json.dumps(benchmark.benchmarker.requests["attributes"]["data"]),
+        ]
+
+        if len(headers) != len(values):
+            raise ValueError("Headers and values length mismatch.")
+
+        return headers, values
+
 
 @GenerativeBenchmarkerOutput.register("html")
 class GenerativeBenchmarkerHTML(GenerativeBenchmarkerOutput):