T-Pro

Parameters

32B

Published

Dec 11, 2024

T-Pro is an AI model developed by T-Bank (Russia), first published in December 2024. It works in the language domain, on tasks such as language modeling/generation, question answering, code generation and mathematical reasoning.

Epoch AI has no training-compute estimate for this model. The model has 32,000,000,000 parameters. It was trained on roughly 142B datapoints.

Access: Open weights (unrestricted). Its weights are openly available. It is built on top of Qwen2.5-32B. Epoch AI rates the confidence of this record as confident.

Full record

Organization: T-Bank
Country of organization: Russia
Domain: Language
Task: Language modeling/generation, Question answering, Code generation, Mathematical reasoning
Compute estimation method: Operation counting
Parameters: 32,000,000,000
Dataset size: 142B
Model accessibility: Open weights (unrestricted)
Open weights: Yes
Base model: Qwen2.5-32B
Epoch confidence: Confident