[Model] Nemotron Parse 1.1 Support (#30864)

Signed-off-by: amitz-nv <203509407+amitz-nv@users.noreply.github.com>
Signed-off-by: Michael Goin <mgoin64@gmail.com>
Co-authored-by: Michael Goin <mgoin64@gmail.com>
This commit is contained in:
amitz-nv
2026-01-05 23:00:14 +02:00
committed by GitHub
parent af1b07b0c5
commit ee21291825
13 changed files with 1117 additions and 31 deletions

View File

@@ -1220,7 +1220,7 @@ class NemotronH_Nano_VL_V2(
n = pixel_values.shape[0]
vit_embeds_list = []
for i in range(0, n, micro_batch_size):
vit_embeds = self.vision_model(pixel_values[i : i + micro_batch_size])
_, vit_embeds = self.vision_model(pixel_values[i : i + micro_batch_size])
vit_embeds = vit_embeds.to(dtype=torch.bfloat16)
h = w = int(vit_embeds.shape[1] ** 0.5)
vit_embeds = vit_embeds.reshape(vit_embeds.shape[0], h, w, -1)
@@ -1695,12 +1695,7 @@ class NemotronH_Nano_VL_V2(
patch_size=patch_size,
norm_mean=hf_config.norm_mean,
norm_std=hf_config.norm_std,
reg_tokens=(
hf_config_vision.args.get("register_multiple")
if hasattr(hf_config_vision, "args")
and isinstance(hf_config_vision.args, dict)
else None
),
**hf_config_vision.args,
)
return RadioModel(config=radio_config)