Qwen3-0.6B – ASR Post-Training Medical French

Snapshot at checkpoint-7000 — mid-training upload before full-dataset restart.

Fine-tuned from Qwen/Qwen3-0.6B on PraxySante/Qwen3-ASR-PostTrain-Complete-Medical-French-Full (~16M medical + oral French sentences: BDPM, PARCOMED, CAS, MORFITT, QUAERO, Wikipedia, FineWeb-2…).

Goal

Correct ASR transcription errors in medical French (drug names, dosages, anatomy, abbreviations).

Downloads last month
-
Safetensors
Model size
0.6B params
Tensor type
BF16
·
Inference Providers NEW
This model isn't deployed by any Inference Provider. 🙋 Ask for provider support

Model tree for PraxySante/Qwen3-0.6B-ASR-PostTrain-Medical-FR

Finetuned
Qwen/Qwen3-0.6B
Finetuned
(952)
this model