Vireo Base inference checkpoint

This repository is an inference-only export. It contains the final model tensor and the architecture configuration needed to construct the model. It does not contain optimizer state, training scripts, launch configuration, logs, experiment-tracking artifacts, provenance bundles, or intermediate checkpoints.

Files

  • checkpoints/final/model.pt: final PyTorch model checkpoint.
  • checkpoints/final/metadata.json: architecture configuration and model-file integrity metadata.

Download

from huggingface_hub import hf_hub_download
import torch

path = hf_hub_download(
    "SaifPunjwani/vireo-base",
    "checkpoints/final/model.pt",
)
checkpoint = torch.load(path, map_location="cpu", weights_only=True)

Construct the Vireo architecture from the config object in checkpoints/final/metadata.json, then load the checkpoint state into that architecture.

Downloads last month

-

Downloads are not tracked for this model. How to track
Inference Providers NEW
This model isn't deployed by any Inference Provider. 🙋 Ask for provider support