newmindai/Mecellem-Qwen3-4B-TR
TEXT GENERATIONConcurrency Cost:1Model Size:4BQuant:BF16Ctx Length:32kPublished:Jan 16, 2026License:apache-2.0Architecture:Transformer0.0K Open Weights Warm

Mecellem-Qwen3-4B-TR is a 4 billion parameter Turkish legal language model developed by newmindai, based on the Qwen3 decoder architecture with a 40,960 token context length. It was continually pre-trained on approximately 270.8 billion tokens of Turkish legal and general web data in a single large-scale phase. This model is specifically optimized for Turkish legal text generation, summarization, and question answering, demonstrating superior performance over its base model in legal quality objectives.

Loading preview...