Text Classification
PEFT
Safetensors
English
biojev
biomedical
qwen3.5
qlora
natural-language-inference
biomedical-nlp
system-one
Instructions to use Gabriel382/BioJev-Nano with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- PEFT
How to use Gabriel382/BioJev-Nano with PEFT:
from peft import PeftModel from transformers import AutoModelForSequenceClassification base_model = AutoModelForSequenceClassification.from_pretrained("Qwen/Qwen3.5-0.8B-Base") model = PeftModel.from_pretrained(base_model, "Gabriel382/BioJev-Nano") - Notebooks
- Google Colab
- Kaggle
File size: 711 Bytes
02030c8 | 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 | # BioJev-Nano — System One API
The System One compatibility server is maintained in the main BioJev repository:
https://github.com/Gabriel382/BioJev
After cloning/installing BioJev, serve this local model directory with:
```bash
python scripts/serve_systemone.py \
--checkpoint /path/to/BioJev-Nano \
--model-name biojev-nano \
--load-in-4bit \
--host 0.0.0.0 \
--port 8000
```
Endpoint:
```text
POST /v1/systemone
```
Supported decision types:
- `choice`
- `noul`
- `score`
BioJev-Nano remains a three-class NLI sequence classifier. The System One endpoint is
a compatibility bridge which converts each candidate into an NLI hypothesis and scores
it using the existing trained checkpoint.
|