// About
What it is
These models are specialized for evaluating sentiment in financial news and market-related content. Unlike general-purpose sentiment models that often misclassify objective financial reporting as neutral, these models are fine-tuned to distinguish subtle positive, negative, and truly neutral signals in financial language.
Each language has its own dedicated model, trained and evaluated on high-quality financial datasets to capture language-specific nuances. This makes them particularly effective for applications such as market sentiment analysis, portfolio risk monitoring, trading signal generation, and financial news aggregation.
// How to use
Drop-in Python snippet
from transformers import pipeline
nlp = pipeline("sentiment-analysis", model="bardsai/finance-sentiment-fr-base")
nlp("Le chiffre d'affaires net a augmente de 30 % pour atteindre 36 millions d'euros."){'label': 'positive', 'score': 0.9999314546585083}// Performance
Evaluation metrics
F1 Macro
0.953
Precision
0.959
Recall
0.949
Accuracy
0.961
// More open source
You might want these too
Open Source
Polish Whisper
Whisper ASR models fine-tuned for Polish Language - winner of Whisper Fine Tuning Event Challenge
Explore the release
Open Source
Jaskier 7b LLM
Top-ranked <=7B LLM (Feb 2024): Fast, smart, open-source. Fine-tuned for exceptional reasoning and instruction-following.
Explore the release
Open Source
Fast Twitter Emotion Detection
Emotion classification model for Polish tweets. Detects joy, optimism, sadness, and anger with high accuracy and speed.
Explore the release
// Need a model like this?
We build production models. Then we open-source the useful ones.