t-tech/T-pro-it-2.0

T-pro-it-2.0 is a 32 billion parameter model developed by t-tech, built upon the Qwen 3 architecture. It incorporates continual pre-training and alignment techniques, with a significant focus on reasoning tasks through its specialized instruction and preference tuning datasets. This model excels in complex reasoning and mathematical problem-solving, offering a notable performance improvement over its base model and other alternatives on Russian language benchmarks.