RefinedNeuro/RN_TR_R2
TEXT GENERATIONConcurrency Cost:1Model Size:8BQuant:FP8Ctx Length:8kLicense:apache-2.0Architecture:Transformer Open Weights Warm

RefinedNeuro/RN_TR_R2 is an 8 billion parameter Turkish-language reasoning model, fine-tuned from ytu-ce-cosmos/Turkish-Llama-8b-DPO-v0.1 using the GRPO method on distilled Q&A data from Qwen3. It specializes in open-ended reasoning tasks in Turkish, particularly excelling in STEM, history, and cultural questions. The model achieves an 82.4% score on the RN_TR_R2_Benchmark_Results, demonstrating significant improvement over its baseline and other 8B models for Turkish reasoning.

Loading preview...