diff --git a/tests/benchmark-models/test_petab_benchmark.py b/tests/benchmark-models/test_petab_benchmark.py
index 0f85926e16..f82787494c 100644
--- a/tests/benchmark-models/test_petab_benchmark.py
+++ b/tests/benchmark-models/test_petab_benchmark.py
@@ -40,6 +40,7 @@
     "Fujita_SciSignal2010",
     # FIXME: re-enable
     "Raia_CancerResearch2011",
+    "Zheng_PNAS2012",
 }
 models = list(sorted(models))
 
@@ -134,7 +135,6 @@ def test_benchmark_gradient(model, scale, sensitivity_method, request):
         "Sneyd_PNAS2002",
         "Bertozzi_PNAS2020",
         "Okuonghae_ChaosSolitonsFractals2020",
-        "Zheng_PNAS2012",
     ):
         # not really worth the effort trying to fix these cases if they
         # only fail on linear scale