t-tech/T-lite-it-2.1

T-lite-it-2.1 is an 8 billion parameter Russian language model developed by t-tech, built upon the Qwen 3 architecture with a 32,768 token context length. It features enhanced instruction following and robust tool-calling capabilities, significantly outperforming its predecessor and Qwen3-8B in agentic applications. Optimized for higher Russian text generation throughput, this model is designed for both general tasks and complex workflows requiring tool integration.

Warm
Public
8B
FP8
32768
License: apache-2.0
Hugging Face

No reviews yet. Be the first to review!