Skip to content
Anna.Mosaki
← Back to projects

04 Sentiment Bench

Financial NLP benchmark: FinBERT vs LSTM vs local LLM — accuracy, latency, and alpha.

🚧

Coming Soon

This demo is still in development.

Status

In Development

Stack

PyTorchFinBERTLLMbacktest

Coming Soon

  • Accuracy, F1, calibration (ECE)
  • Latency and cost per 1k docs
  • Downstream long/short Sharpe comparison