Nejatgar
Ajouter un avis SuivezVue d'ensemble
-
Date de création novembre 12, 2022
-
Secteur Gestion
-
Offres d'emploi 0
-
Consultés 131
Company Description
DeepSeek’s First-generation Reasoning Models
DeepSeek’s first-generation thinking designs, accomplishing performance comparable to OpenAI-o1 across math, code, and .
Models
DeepSeek-R1
Distilled models
DeepSeek team has demonstrated that the reasoning patterns of bigger designs can be distilled into smaller sized designs, resulting in much better efficiency compared to the reasoning patterns found through RL on small models.
Below are the models produced through fine-tuning versus numerous thick models extensively used in the research community utilizing thinking data generated by DeepSeek-R1. The evaluation results demonstrate that the distilled smaller sized dense designs carry out exceptionally well on standards.
DeepSeek-R1-Distill-Qwen-1.5 B
DeepSeek-R1-Distill-Qwen-7B
DeepSeek-R1-Distill-Llama-8B
DeepSeek-R1-Distill-Qwen-14B
DeepSeek-R1-Distill-Qwen-32B
DeepSeek-R1-Distill-Llama-70B
License
The design weights are licensed under the MIT License. DeepSeek-R1 series support business usage, permit for any adjustments and acquired works, consisting of, but not limited to, distillation for training other LLMs.
