Skip to content

nli:modernbert-large

3 TagsUpdated 396M params512 contextEnglishMITby Moritz Laurer

ModernBERT-large, Apache-2.0. Faster, slightly less accurate (0.515).

396m
ollaya run nli:modernbert-large --preset triage "I was charged twice for my subscription this month and want a refund."

Details

  • graphda223fdbe719 · 4 MBonnx · ModernBERT-large-zeroshot-v2.0 · 396M · fp32
  • weights9e88e3ba0721 · 792 MBhuggingface.co/MoritzLaurer/ModernBERT-large-zeroshot-v2.0/resolve/a51e07b…/model.safetensors
  • tokenizer6c8aaa9a5420 · 4 MBhuggingface.co/MoritzLaurer/ModernBERT-large-zeroshot-v2.0/resolve/a51e07b…/tokenizer.json
  • decisione47095bd7afa · 1 KB{"engine": "onnx", "family": "nli", "encoder": "MoritzLaurer/ModernBERT-large-zeroshot-v2.0", "layout": "nli-pairs-v1", …}
  • calibration316915b3e458 · 84 B{"temperature": [1.0, 1.0, 1.0]}
  • license445805446407 · 10 KBModernBERT-large zero-shot v2.0 by Moritz Laurer (https://huggingface.co/MoritzLaurer/ModernBERT-large-zeroshot-v2.0)

Every layer is checked against its sha256 when it is pulled. Weights and tokenizers download from the model author's Hugging Face repository at a pinned commit; Ollaya never re-hosts them.

NLI zero-shot readme and all models