Token Classification
Transformers
Safetensors
English
Vietnamese
gemma4
image-text-to-text
systemone
extraction-model
text-to-typed-output
information-extraction
key-information-extraction
extractive-qa
span-extraction
structured-output
document-ai
multilingual
Instructions to use beyoru/gemmaExtract with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use beyoru/gemmaExtract with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("token-classification", model="beyoru/gemmaExtract")# pip install -U transformers accelerate # Load model directly from transformers import AutoProcessor, AutoModelForMultimodalLM processor = AutoProcessor.from_pretrained("beyoru/gemmaExtract") model = AutoModelForMultimodalLM.from_pretrained("beyoru/gemmaExtract", device_map="auto") - Notebooks
- Google Colab
- Kaggle
Note: text input only
Browse files
README.md
CHANGED
|
@@ -30,6 +30,8 @@ You give gemmaExtract a document and a list of fields you want. It returns each
|
|
| 30 |
document ──► one forward pass ──► {field: value, span, p_null, confidence}
|
| 31 |
```
|
| 32 |
|
|
|
|
|
|
|
| 33 |
## Quick start
|
| 34 |
|
| 35 |
```python
|
|
@@ -132,6 +134,7 @@ Measured on half of an H200, bf16, batch 1, default null threshold.
|
|
| 132 |
- **Span-based**: answers that must be paraphrased or merged from scattered places (long free-form answers, multi-line addresses) are weaker than with a generative model.
|
| 133 |
- Up to **2,048 tokens per pass**; longer documents are truncated (`usage.truncated` tells you).
|
| 134 |
- Languages other than English and Vietnamese work (see the XQuAD table) but absent-field detection has only been checked in English and Vietnamese.
|
|
|
|
| 135 |
- Not supported yet: `time`, `duration`, `estimate`, `enum` fields and image input.
|
| 136 |
- Runs with `transformers` (≥ 5.17); there is no serving-engine integration yet.
|
| 137 |
|
|
|
|
| 30 |
document ──► one forward pass ──► {field: value, span, p_null, confidence}
|
| 31 |
```
|
| 32 |
|
| 33 |
+
> **Text input only.** gemmaExtract currently reads plain text (`state` must be a string). For scanned documents, photos or PDFs, run OCR first and pass the recognized text. Image input is not supported yet.
|
| 34 |
+
|
| 35 |
## Quick start
|
| 36 |
|
| 37 |
```python
|
|
|
|
| 134 |
- **Span-based**: answers that must be paraphrased or merged from scattered places (long free-form answers, multi-line addresses) are weaker than with a generative model.
|
| 135 |
- Up to **2,048 tokens per pass**; longer documents are truncated (`usage.truncated` tells you).
|
| 136 |
- Languages other than English and Vietnamese work (see the XQuAD table) but absent-field detection has only been checked in English and Vietnamese.
|
| 137 |
+
- Text input only: images, scans and PDFs need an OCR step first.
|
| 138 |
- Not supported yet: `time`, `duration`, `estimate`, `enum` fields and image input.
|
| 139 |
- Runs with `transformers` (≥ 5.17); there is no serving-engine integration yet.
|
| 140 |
|