H 4 Research
Add a review FollowOverview
-
Founded Date October 9, 2012
-
Sectors Education
-
Posted Jobs 0
-
Viewed 42
Company Description
DeepSeek’s First-generation Reasoning Models
DeepSeek’s first-generation thinking designs, attaining performance equivalent to OpenAI-o1 throughout mathematics, code, and reasoning jobs.
Models
DeepSeek-R1
Distilled models
DeepSeek team has demonstrated that the of larger models can be distilled into smaller designs, resulting in better efficiency compared to the thinking patterns discovered through RL on small models.
Below are the models produced by means of fine-tuning against a number of thick designs widely utilized in the research community utilizing thinking information produced by DeepSeek-R1. The evaluation results demonstrate that the distilled smaller dense models perform 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 model weights are certified under the MIT License. DeepSeek-R1 series support business usage, permit any adjustments and derivative works, consisting of, however not limited to, distillation for training other LLMs.