The TempestLLM model collection consists of the base foundation model, an instruction fine-tuned (SFT) version, and other related variants.
AI & ML interests
Our goal is to train the largest possible SSM model while minimizing infrastructure requirements. This approach reduces both economic and environmental impact without significantly compromising the model's linguistic performance.
Recent Activity
A suite of models and a dataset designed to assess the quality of natural language and code data.
-
TempestTeam/EuroBERT-210m-Quality
Text Classification • 0.2B • Updated • 11 • 1 -
TempestTeam/EuroBERT-210m-Quality-NL
Text Classification • 0.2B • Updated • 7 • 1 -
TempestTeam/EuroBERT-210m-Quality-CL
Text Classification • 0.2B • Updated • 3 • 1 -
TempestTeam/dataset-quality
Viewer • Updated • 44.8k • 14 • 1
The TempestLLM model collection consists of the base foundation model, an instruction fine-tuned (SFT) version, and other related variants.
A suite of models and a dataset designed to assess the quality of natural language and code data.
-
TempestTeam/EuroBERT-210m-Quality
Text Classification • 0.2B • Updated • 11 • 1 -
TempestTeam/EuroBERT-210m-Quality-NL
Text Classification • 0.2B • Updated • 7 • 1 -
TempestTeam/EuroBERT-210m-Quality-CL
Text Classification • 0.2B • Updated • 3 • 1 -
TempestTeam/dataset-quality
Viewer • Updated • 44.8k • 14 • 1