Add Nemotron 3 Nano tests (#18119)

Signed-off-by: Shahar Mor <smor@nvidia.com>

test/lm_eval_configs/NVIDIA-Nemotron-3-Nano-30B-A3B-BF16.yaml

@@ -0,0 +1,13 @@
+model_name: "nvidia/NVIDIA-Nemotron-3-Nano-30B-A3B-BF16"
+tasks:
+- name: "gsm8k"
+  metrics:
+  - name: "exact_match,strict-match"
+    value: 0.847
+  - name: "exact_match,flexible-extract"
+    value: 0.556
+limit: 1319
+num_concurrent: 128
+num_fewshot: 5
+apply_chat_template: false
+fewshot_as_multiturn: true

test/lm_eval_configs/NVIDIA-Nemotron-3-Nano-30B-A3B-FP8.yaml

@@ -0,0 +1,13 @@
+model_name: "nvidia/NVIDIA-Nemotron-3-Nano-30B-A3B-FP8"
+tasks:
+- name: "gsm8k"
+  metrics:
+  - name: "exact_match,strict-match"
+    value: 0.847
+  - name: "exact_match,flexible-extract"
+    value: 0.556
+limit: 1319
+num_concurrent: 128
+num_fewshot: 5
+apply_chat_template: false
+fewshot_as_multiturn: true

test/registered/models/test_nvidia_nemotron_3_nano.py

@@ -0,0 +1,41 @@
+import unittest
+
+from sglang.test.ci.ci_register import register_cuda_ci
+from sglang.test.kits.lm_eval_kit import LMEvalMixin
+from sglang.test.server_fixtures.default_fixture import DefaultServerBase
+
+register_cuda_ci(est_time=180, suite="stage-b-test-large-2-gpu")
+
+NEMOTRON_3_NANO_THINKING_ARGS = [
+    "--trust-remote-code",
+    "--tool-call-parser",
+    "qwen3_coder",
+    "--reasoning-parser",
+    "deepseek-r1",
+]
+
+
+class TestNvidiaNemotron3Nano30BBF16(LMEvalMixin, DefaultServerBase):
+    """Test Nemotron-3-Nano-30B BF16 model with lm-eval GSM8K evaluation."""
+
+    model = "nvidia/NVIDIA-Nemotron-3-Nano-30B-A3B-BF16"
+    model_config_name = "lm_eval_configs/NVIDIA-Nemotron-3-Nano-30B-A3B-BF16.yaml"
+    other_args = [
+        "--tp-size",
+        "2",
+    ] + NEMOTRON_3_NANO_THINKING_ARGS
+
+
+class TestNvidiaNemotron3Nano30BFP8(LMEvalMixin, DefaultServerBase):
+    """Test Nemotron-3-Nano-30B FP8 model with lm-eval GSM8K evaluation."""
+
+    model = "nvidia/NVIDIA-Nemotron-3-Nano-30B-A3B-FP8"
+    model_config_name = "lm_eval_configs/NVIDIA-Nemotron-3-Nano-30B-A3B-FP8.yaml"
+    other_args = [
+        "--tp-size",
+        "2",
+    ] + NEMOTRON_3_NANO_THINKING_ARGS
+
+
+if __name__ == "__main__":
+    unittest.main()