hishab/titulm-llama-3.2-1b-v1.0
TEXT GENERATIONConcurrency Cost:1Model Size:1BQuant:BF16Ctx Length:32kPublished:Oct 3, 2024License:llama3.2Architecture:Transformer0.0K Warm

The hishab/titulm-llama-3.2-1b-v1.0 model is a 1.23 billion parameter Llama 3.2 architecture, continually pre-trained by Hishab on 6 billion tokens of extensive Bangla datasets. This model is specifically optimized for high-quality Bangla text generation and understanding, demonstrating superior performance in Bangla language evaluation benchmarks. While primarily focused on Bengali, it also retains secondary English language capabilities, making it suitable for applications requiring robust Bangla language processing.

Loading preview...