Instructions to use reachjalil/gemma-4-audio-encoder-teacher with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use reachjalil/gemma-4-audio-encoder-teacher with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("feature-extraction", model="reachjalil/gemma-4-audio-encoder-teacher")# Load model directly from transformers import AutoModel model = AutoModel.from_pretrained("reachjalil/gemma-4-audio-encoder-teacher", dtype="auto") - Notebooks
- Google Colab
- Kaggle
Gemma 4 Audio Encoder Teacher
Public derived teacher package for the Gemma 4 audio encoder teacher surface used by Waker-style registry repos.
Artifact layout
This repo contains a packaged teacher artifact with:
package.manifest.jsonapp-contract.jsonconsumer.example.jsonmodel/
The intended local bootstrap path in downstream repos is:
data/assets/teachers/gemma-4-audio-encoder-teacher/package.manifest.json
Source model
Derived from google/gemma-4-E2B-it.
Notes
This artifact is meant to be downloaded as a teacher/bootstrap dependency. Projection artifacts are not included in this v1 bootstrap package.