biondizzle/vllm
1
0
Fork 0
Code Issues Pull Requests Actions 2 Packages Projects Releases Wiki Activity

[Model] rename NemotronH_Nano_VL -> NemotronH_Nano_VL_V2 (#25708)

Browse Source 
Signed-off-by: Tomer Asida <57313761+tomeras91@users.noreply.github.com>
This commit is contained in:
tomeras91
2025-09-26 02:10:29 +03:00
committed by GitHub
parent 8c435c9bce
commit 57329a8c01
3 changed files with 6 additions and 6 deletions
Download Patch File Download Diff File Expand all files Collapse all files

2
tests/models/registry.py
View File

@@ -531,7 +531,7 @@ _MULTIMODAL_EXAMPLE_MODELS = {
trust_remote_code=True), trust_remote_code=True),
"Llama_Nemotron_Nano_VL" : _HfExamplesInfo("nvidia/Llama-3.1-Nemotron-Nano-VL-8B-V1", # noqa: E501 "Llama_Nemotron_Nano_VL" : _HfExamplesInfo("nvidia/Llama-3.1-Nemotron-Nano-VL-8B-V1", # noqa: E501
trust_remote_code=True), trust_remote_code=True),
"NemotronH_Nano_VL": _HfExamplesInfo("nano_vl_dummy", "NemotronH_Nano_VL_V2": _HfExamplesInfo("nano_vl_dummy",
is_available_online=False, is_available_online=False,
trust_remote_code=True), trust_remote_code=True),
"Ovis": _HfExamplesInfo("AIDC-AI/Ovis2-1B", trust_remote_code=True, "Ovis": _HfExamplesInfo("AIDC-AI/Ovis2-1B", trust_remote_code=True,

8
vllm/model_executor/models/nano_nemotron_vl.py
View File

@@ -869,8 +869,8 @@ class NanoNemotronVLDummyInputsBuilder(
info=NanoNemotronVLProcessingInfo, info=NanoNemotronVLProcessingInfo,
dummy_inputs=NanoNemotronVLDummyInputsBuilder, dummy_inputs=NanoNemotronVLDummyInputsBuilder,
) )
class NemotronH_Nano_VL(nn.Module, HasInnerState, IsHybrid, class NemotronH_Nano_VL_V2(nn.Module, HasInnerState, IsHybrid,
SupportsMultiModal): SupportsMultiModal):
@classmethod @classmethod
def get_placeholder_str(cls, modality: str, i: int) -> Optional[str]: def get_placeholder_str(cls, modality: str, i: int) -> Optional[str]:
@@ -1249,7 +1249,7 @@ class NemotronH_Nano_VL(nn.Module, HasInnerState, IsHybrid,
try: try:
print("=" * 100) print("=" * 100)
print("NemotronH_Nano_VL Model Architecture") print("NemotronH_Nano_VL_V2 Model Architecture")
print("=" * 100) print("=" * 100)
total_params = 0 total_params = 0
@@ -1333,7 +1333,7 @@ class NemotronH_Nano_VL(nn.Module, HasInnerState, IsHybrid,
component_info[component]["size"] += param.numel() component_info[component]["size"] += param.numel()
return { return {
"model_name": "NemotronH_Nano_VL", "model_name": "NemotronH_Nano_VL_V2",
"total_parameters": total_params, "total_parameters": total_params,
"memory_estimate_mb": total_params * 2 / (1024**2), # bfloat16 "memory_estimate_mb": total_params * 2 / (1024**2), # bfloat16
"components": component_info, "components": component_info,

2
vllm/model_executor/models/registry.py
View File

@@ -231,7 +231,7 @@ _MULTIMODAL_MODELS = {
"GraniteSpeechForConditionalGeneration": ("granite_speech", "GraniteSpeechForConditionalGeneration"), # noqa: E501 "GraniteSpeechForConditionalGeneration": ("granite_speech", "GraniteSpeechForConditionalGeneration"), # noqa: E501
"H2OVLChatModel": ("h2ovl", "H2OVLChatModel"), "H2OVLChatModel": ("h2ovl", "H2OVLChatModel"),
"InternVLChatModel": ("internvl", "InternVLChatModel"), "InternVLChatModel": ("internvl", "InternVLChatModel"),
"NemotronH_Nano_VL": ("nano_nemotron_vl", "NemotronH_Nano_VL"), "NemotronH_Nano_VL_V2": ("nano_nemotron_vl", "NemotronH_Nano_VL_V2"),
"InternS1ForConditionalGeneration": ("interns1", "InternS1ForConditionalGeneration"), # noqa: E501 "InternS1ForConditionalGeneration": ("interns1", "InternS1ForConditionalGeneration"), # noqa: E501
"InternVLForConditionalGeneration": ("interns1", "InternS1ForConditionalGeneration"), # noqa: E501 "InternVLForConditionalGeneration": ("interns1", "InternS1ForConditionalGeneration"), # noqa: E501
"Idefics3ForConditionalGeneration":("idefics3","Idefics3ForConditionalGeneration"), "Idefics3ForConditionalGeneration":("idefics3","Idefics3ForConditionalGeneration"),