Downloads · 30 days
0
span-marker-xlm-roberta-base-fewnerd-onnx
Berk/span-marker-xlm-roberta-base-fewnerd-onnx
span-marker-xlm-roberta-base-fewnerd-onnx is a token classification model from Berk. Use it when you need labels on individual words, such as names. It is set up for onnxruntime. The card lists the license as cc-by-sa-4.0.
ONNX export of tomaarsen/span-marker-xlm-roberta-base-fewnerd-fine-super for browser-based inference.
Access
Public
Updated May 12, 2026
Repo size
296 MB
Likes
0
Public
Repo makeup
Click a slice to open those files.
.onnx279 MB · 94%
At a glance
- Task
- Token Classification
- Library
- onnxruntime
- License
- cc-by-sa-4.0
- Access
- Public
- Created
- May 12, 2026
- Updated
- May 12, 2026
- SHA
- 09b56911
Try a prompt
Base models
Datasets
Model card
From the Hugging Face model README
SpanMarker XLM-RoBERTa Base FewNERD (ONNX)
ONNX export of tomaarsen/span-marker-xlm-roberta-base-fewnerd-fine-super for browser-based inference.
Model Details
- Base model: XLM-RoBERTa Base (multilingual, 10+ languages)
- Training data: FewNERD fine-grained supervised
- Entity types: 66 types + O (including
building-hotel,building-airport,location-GPE,location-park,location-bodiesofwater,location-mountain, etc.) - Quantization: INT8 dynamic quantization on encoder
Files
| File | Size | Description |
|---|---|---|
encoder.onnx | ~266 MB | XLM-RoBERTa encoder (INT8 quantized) |
classifier.onnx | 0.4 MB | Linear classifier (1536 -> 67 classes) |
labels.json | 2 KB | ID to label mapping |
tokenizer.json | 17 MB | SentencePiece tokenizer |
Inference
SpanMarker uses a custom architecture:
- Tokenize input and enumerate all spans (up to
entity_max_length=8tokens) - Insert start/end marker tokens into the sequence
- Run encoder to get hidden states
- Extract start marker embeddings from
[start_idx : start_idx + n_pairs] - Extract end marker embeddings from
[start_idx + n_pairs : start_idx + n_pairs*2] - Concatenate start+end -> (n_pairs, 1536) features
- Pad to (256, 1536) and run classifier -> (256, 67) logits
- Softmax + filter non-O predictions above threshold
Limitations
- Fixed sequence length of 512 (batch=1)
- Longer texts need to be chunked (the original model does this internally with batching)
- INT8 quantization slightly reduces scores (~0.05-0.1 lower) but maintains correct predictions
License
CC-BY-SA-4.0 (same as original model)
- Task
- Token Classification
- Library
- onnxruntime
- License
- cc-by-sa-4.0
- Languages
- en, multilingual
- Created
- May 12, 2026
- Updated
- May 12, 2026