From e42c230e4ff4e3b44b100ced7d731a5190948d2e Mon Sep 17 00:00:00 2001
From: Turner Jabbour <doubleujabbour@gmail.com>
Date: Sun, 2 Aug 2026 20:26:48 -0600
Subject: [PATCH] [Bugfix] serving_llama70B_tp4 benchmark was silently running
 at tensor_parallel_size=1 (#50766)

Signed-off-by: Turner Jabbour <doubleujabbour@gmail.com>
Co-authored-by: Claude <noreply@anthropic.com>
---
 .buildkite/performance-benchmarks/tests/serving-tests.json | 1 +
 1 file changed, 1 insertion(+)

diff --git a/.buildkite/performance-benchmarks/tests/serving-tests.json b/.buildkite/performance-benchmarks/tests/serving-tests.json
index dabc218399..ac0eb213b9 100644
--- a/.buildkite/performance-benchmarks/tests/serving-tests.json
+++ b/.buildkite/performance-benchmarks/tests/serving-tests.json
@@ -76,6 +76,7 @@
       "test_name": "serving_llama70B_tp4_random_128_128",
       "server_parameters": {
         "model": "meta-llama/Llama-3.3-70B-Instruct",
+        "tensor_parallel_size": 4,
         "async_scheduling": "",
 	"no_enable_prefix_caching": "",
         "max_num_batched_tokens": 8192