diff --git a/python/sglang/test/accuracy_test_runner.py b/python/sglang/test/accuracy_test_runner.py
index c35dba6497b4..abe8cc3d59c1 100644
--- a/python/sglang/test/accuracy_test_runner.py
+++ b/python/sglang/test/accuracy_test_runner.py
@@ -29,6 +29,7 @@ class AccuracyTestParams:
     top_p: Optional[float] = None
     top_k: Optional[int] = None
     repeat: Optional[int] = None
+    api: Optional[str] = None  # "chat" or "completion"; defaults to "chat" in run_eval
 
 
 @dataclass
@@ -86,6 +87,7 @@ def _run_simple_eval(
     top_p: Optional[float] = None,
     top_k: Optional[int] = None,
     repeat: Optional[int] = None,
+    api: Optional[str] = None,
 ) -> Tuple[bool, Optional[str], Optional[dict]]:
     """Run evaluation using simple_eval backend (run_eval.py).
 
@@ -110,6 +112,9 @@ def _run_simple_eval(
             num_threads=num_threads or 1024,
         )
 
+        if api is not None:
+            args.api = api
+
         if max_tokens is not None:
             args.max_tokens = max_tokens
 
@@ -482,6 +487,7 @@ def run_accuracy_test(
             top_p=params.top_p,
             top_k=params.top_k,
             repeat=params.repeat,
+            api=params.api,
         )
 
     if not success:
diff --git a/test/registered/backends/test_flashinfer_trtllm_gen_moe_backend.py b/test/registered/backends/test_flashinfer_trtllm_gen_moe_backend.py
index b63447a60cd5..76b47ffd4dee 100644
--- a/test/registered/backends/test_flashinfer_trtllm_gen_moe_backend.py
+++ b/test/registered/backends/test_flashinfer_trtllm_gen_moe_backend.py
@@ -59,7 +59,7 @@ def test_gsm8k(self):
         )
         metrics = run_eval(args)
         print(f"{metrics=}")
-        self.assertGreater(metrics["score"], 0.93)
+        self.assertGreater(metrics["score"], 0.89)
 
 
 class FlashinferTrtllmGenMoeBackendBF16Base:
diff --git a/test/registered/perf/test_dpsk_r1_fp4_4gpu_perf.py b/test/registered/perf/test_dpsk_r1_fp4_4gpu_perf.py
index b03c34337d26..23714a8ce1fc 100644
--- a/test/registered/perf/test_dpsk_r1_fp4_4gpu_perf.py
+++ b/test/registered/perf/test_dpsk_r1_fp4_4gpu_perf.py
@@ -62,7 +62,10 @@ def test_deepseek_r1_fp4_all_variants(self):
             models=variants,
             test_name="DeepSeek-R1-0528-NVFP4-v2 Unified",
             accuracy_params=AccuracyTestParams(
-                dataset="gsm8k", baseline_accuracy=0.935
+                dataset="gsm8k",
+                baseline_accuracy=0.935,
+                num_examples=200,
+                api="completion",
             ),
             performance_params=PerformanceTestParams(
                 profile_dir="performance_profiles_deepseek_r1_fp4",