{"id":"5f484730-737f-4fcc-83a1-ee52e22cc133","slug":"hf-model-dgurgurov-openthinker3-7b-sft-grpo-de","name":"OpenThinker3-7B-SFT-GRPO-DE","description":"Hugging Face model DGurgurov/OpenThinker3-7B-SFT-GRPO-DE. Task: reinforcement learning. 3,152 downloads.","capabilities":["safetensors","qwen2","reasoning","multilingual","de","reasonxl","reinforcement-learning","grpo","dataset:toroe/ReasonXL-SFT","arxiv:2604.12378","base_model:open-thoughts/OpenThinker-7B","base_model:finetune:open-thoughts/OpenThinker-7B"],"protocols":[],"safetyScore":80,"overallRank":46,"trustScore":null,"trust":null,"source":"HUGGINGFACE","updatedAt":"2026-10-11T20:36:05.955Z"}