tartuNLP/llama-estllm-prototype-0825
TEXT GENERATIONConcurrency Cost:1Model Size:8BQuant:FP8Ctx Length:32kPublished:Sep 1, 2025License:llama3.1Architecture:Transformer0.0K Cold

The tartuNLP/llama-estllm-prototype-0825 is an 8 billion parameter instruction-following causal language model developed by TartuNLP and TalTechNLP. It is a prototype version of EstLLM, continuously pre-trained on 35B tokens from Llama-3.1-8B, with a focus on enhancing Estonian language capabilities. This model excels in Estonian language competence and instruction-following tasks, serving as a baseline for future improvements in multilingual LLMs.

Loading preview...