watt-flash-0.1

Code: github.com/zaitgeist-charly/zaitlabs-watt-flash-0.1

State + typed questions (noul, choice, score) -> option probabilities. One forward pass per request.

git clone https://github.com/zaitgeist-charly/zaitlabs-watt-flash-0.1 && cd zaitlabs-watt-flash-0.1 && uv sync
uv run python -m watt_flash '{"state": "Doppelt belastet, bitte erstatten.", "questions": {"refund": {"type": "noul", "instructions": "Refund requested?"}}}'

mmBERT-small backbone, 140M parameters, 8,192-token context. Multilingual question corpus, 490,000 examples, 20 languages.

Downloads last month

-

Downloads are not tracked for this model. How to track
Safetensors
Model size
0.1B params
Tensor type
BF16
·
Inference Providers NEW
This model isn't deployed by any Inference Provider. 🙋 Ask for provider support

Model tree for zaitlabs/watt-flash-0.1

Finetuned
(52)
this model