Create a Model FileSet#

Create a FileSet containing your base model checkpoint before creating a Model Entity.

Prerequisites#

Obtained the base URL of your NeMo Platform.
For HuggingFace models: Created a secret with your HF token. Refer to Manage Secrets.
Set the NMP_BASE_URL environment variable.

export NMP_BASE_URL="https://your-nemo-platform-url"

From HuggingFace Hub#

The most common method is downloading directly from HuggingFace:

import os
from nemo_platform import NeMoPlatform
from nemo_platform._exceptions import ConflictError
from nemo_platform.types.files import HuggingfaceStorageConfigParam

client = NeMoPlatform(
    base_url=os.environ.get("NMP_BASE_URL", "http://localhost:8080"),
    workspace="default",
)

HF_REPO_ID = "meta-llama/Llama-3.2-1B-Instruct"
MODEL_NAME = "llama-3-2-1b"
HF_SECRET_NAME = "my-hf-token"

# First, create a secret for your HF token (if not already created)
try:
    hf_secret = client.secrets.create(
        name=HF_SECRET_NAME,
        workspace="default",
        data="hf_your_token_here"  # Replace with your actual HuggingFace token
    )
    print(f"Created secret: {HF_SECRET_NAME}")
except ConflictError:
    print(f"Secret '{HF_SECRET_NAME}' already exists, continuing...")
    hf_secret = client.secrets.retrieve(name=HF_SECRET_NAME, workspace="default")

# Create FileSet from HuggingFace
try:
    fileset = client.files.filesets.create(
        workspace="default",
        name=MODEL_NAME,
        description="Llama 3.2 1B Instruct from HuggingFace",
        storage=HuggingfaceStorageConfigParam(
            type="huggingface",
            repo_id=HF_REPO_ID,
            repo_type="model",
            token_secret=hf_secret.name
        )
    )
    print(f"Created FileSet: {fileset.name}")
except ConflictError:
    print(f"FileSet '{MODEL_NAME}' already exists, retrieving...")
    fileset = client.files.filesets.retrieve(workspace="default", name=MODEL_NAME)

print(f"FileSet ready: {fileset.name}")

Tip

For gated models (like Llama), you need to:

Accept the model license on HuggingFace
Create a HuggingFace token with read access
Store the token as a secret in the platform

From NGC#

For models from NVIDIA NGC:

import os
from nemo_platform import NeMoPlatform
from nemo_platform._exceptions import ConflictError
from nemo_platform.types.files import NGCStorageConfigParam

client = NeMoPlatform(
    base_url=os.environ.get("NMP_BASE_URL", "http://localhost:8080"),
    workspace="default",
)

# Create FileSet from NGC
try:
    fileset = client.files.filesets.create(
        workspace="default",
        name="nemotron-mini-4b",
        description="Nemotron Mini 4B from NGC",
        storage=NGCStorageConfigParam(
            type="ngc",
            org="nvidia",
            team="nemo",
            resource="nemotron-mini-4b-instruct",  # NGC resource name
            version="1.0",
            api_key_secret="my-ngc-key"  # Secret containing NGC API key
        )
    )
    print(f"Created FileSet: {fileset.name}")
except ConflictError:
    print("FileSet already exists, retrieving...")
    fileset = client.files.filesets.retrieve(workspace="default", name="nemotron-mini-4b")

Check FileSet Status#

Files are downloaded in the background after you create a FileSet. Check the status:

# List files in the FileSet
files = client.files.list(
    workspace="default",
    fileset="llama-3-2-1b",
)

print(f"Files in FileSet ({len(files)} total):")
for f in files[:10]:  # Show first 10
    print(f"  - {f.path} ({f.size:,} bytes)")

Next Steps#

Proceed to create a Model Entity.