model: Support Hybrid Mamba2 NemotronHForCausalLM (nvidia/NVIDIA-Nemotron-Nano-9B-v2) (#10909)

Signed-off-by: Netanel Haber <nhaber@nvidia.com>
2025-10-08 19:37:38 +03:00
parent c882b5ae75
commit d6837aea4d
35 changed files with 3280 additions and 854 deletions
--- a/test/srt/models/test_generation_models.py
+++ b/test/srt/models/test_generation_models.py
@@ -91,6 +91,11 @@ ALL_MODELS = [
        trust_remote_code=True,
        skip_long_prompt=True,
    ),
+    ModelCase(
+        "nvidia/NVIDIA-Nemotron-Nano-9B-v2",
+        trust_remote_code=True,
+        skip_long_prompt=True,
+    ),
    ModelCase(
        "swiss-ai/Apertus-8B",
        trust_remote_code=True,