Fill-Mask
Transformers
Safetensors
nandi_encoder
encoder
masked-language-modeling
indic
multilingual
code
long-context
frontiersmind
nandi
custom_code
Instructions to use FrontiersMind/NandiBERT-450M with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use FrontiersMind/NandiBERT-450M with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("fill-mask", model="FrontiersMind/NandiBERT-450M", trust_remote_code=True)# pip install -U transformers accelerate # Load model directly from transformers import AutoModelForMaskedLM model = AutoModelForMaskedLM.from_pretrained("FrontiersMind/NandiBERT-450M", trust_remote_code=True, device_map="auto") - Notebooks
- Google Colab
- Kaggle
Commit History
Yesterday
-
Model card: note on supported Indic languages
Verified signatureb83d4d1 -
Model card: note that results are the average of 3 seeds
Verified signature2be56a6 -
Model card: normal citation (remove placeholder note)
Verified signature595c97a -
Model card: bold Indic results, Indic average, launch blog link, technical report coming soon
Verified signatured07319c -
Parameter counts (445M encoder, 146M for the 150M); remove internal references from code and config
Verified signature5eca2f3 -
Classifiers use the pretrained MLM head; tokenizer word spaces; README tokenizer needs trust_remote_code
Verified signatureSequence and token classification put the pretrained dense MLM head (dense, act, RMSNorm) before the new linear layer when config.classifier_head is "mlm" (now the default). classifier_head=None keeps the plain setup. is_split_into_words=True gives every word its leading space. README: AutoTokenizer needs trust_remote_code=True, plus a short fine-tuning section.4acb258 -
Model cards: use full names with size for compared models
Verified signaturefa52da1 -
Add English benchmark results
Verified signature36e1db2
2 days ago
-
Add Indic benchmark results
Verified signature3b189d0 -
Update model card: Indic benchmarks first, remove MASSIVE Intent, parameter counts
Verified signature53dbc5c -
Add final checkpoint (model, config, tokenizer, remote code)
Verified signature66f3476
3 days ago
-
Update README.md
Verified signaturee13e5c2 -
Upload nandibert-banner_1.svg
Verified signature652f459 -
Update README.md
Verified signature4e45888 -
initial commit
Verified signaturecf5d5c3