bnicenboim/gpt2-spanish-onnx

gpt2-spanish (ONNX)

ONNX export of DeepESP/gpt2-spanish for use in the browser with Transformers.js.

Usage with Transformers.js

import { pipeline } from '@huggingface/transformers';

const generator = await pipeline(
  'text-generation',
  'bnicenboim/gpt2-spanish-onnx'
);

const result = await generator('El gato estaba en', { max_new_tokens: 50 });
console.log(result[0].generated_text);

Model details

• Base model: DeepESP/gpt2-spanish
• Architecture: GPT-2 (~117M parameters)
• Language: Spanish (es)
• Format: ONNX (fp32) + quantized int8
• Exported with: 🤗 Optimum
• Training corpus: 11.5GB of Spanish text including Wikipedia, novels, short stories, theater, poetry, and essays

Intended use

Causal language modelling / text completion in Spanish, running client-side in the browser via Transformers.js. Not an instruction or chat model.