# Andoni Sudupe: Open-Weight Models and Datasets
Source: https://savrn.com/model-publishers/ansu
Markdown alternate of the page above; the site index is https://savrn.com/llms.txt

---

## Models

Model

### [whisper-large-v3-basque](https://savrn.com/models/whisper-large-v3-basque)

[Andoni Sudupe](https://savrn.com/model-publishers/ansu)

This model is a fine-tuned version of openai/whisper-large-v3 on the None dataset. It achieves the following results on the evaluation set: The following hyperparameters were used during training: - learningrate: 1e-05 - trainbatchsize: 256 - evalbatchsize: 32 - distributedtype: multi-GPU - numdevices: 2 - totaltrainbatchsize: 512 - totalevalbatchsize: 64 - lrschedulertype: linear - lrschedulerwarmupsteps: 500 - trainingsteps: 5000 - mixedprecisiontraining: Native AMP - Transformers 4.38.0 - Pytorch 2.1.1+cu121 - Datasets 2.8.0 - Tokenizers 0.15.2

Open weights apache-2.0 1.6B parameters

[View model](https://savrn.com/models/whisper-large-v3-basque)

## Explore More

- [All model publishers](https://savrn.com/model-publishers)
- [The model directory](https://savrn.com/models)
- [The dataset directory](https://savrn.com/datasets)

## Source

- Listed from their public repositories, read 2026-09-30.
- [Hugging Face profile](https://huggingface.co/Ansu)
- [How the hub is built](https://savrn.com/model-hub/methodology)