Modelpublic

zenlm/zen3-asr-0.6B

sourceHugging Faceapache-2.0updated 2mo agoView on Hugging Face

0likes7downloads

Model Card

Zen3 ASR 0.6B

Compact ~0.6B Zen3 automatic speech recognition model, sized for edge devices and latency-sensitive transcription.

Derived by fine-tuning `Qwen/Qwen3-ASR-0.6B` (Alibaba Cloud, Apache-2.0).

—Architecture: Qwen3ASRForConditionalGeneration (qwen3_asr)
—Parameters: ~0.6B
—Base model: `Qwen/Qwen3-ASR-0.6B`

Weights

This repository contains the model weights: model.safetensors, config and tokenizer files.

The model uses the qwen3_asr architecture and loads with transformers (>= 4.57). It is API-compatible with the upstream base — follow the inference recipe on the base model card `Qwen/Qwen3-ASR-0.6B`.

Provenance

Fine-tuned from `Qwen/Qwen3-ASR-0.6B` (Apache-2.0). See NOTICE for full attribution.