commit
9153fe5461
1 changed files with 2 additions and 0 deletions
@ -0,0 +1,2 @@ |
|||||
|
<br>DeepSeek open-sourced DeepSeek-R1, an LLM fine-tuned with reinforcement learning (RL) to enhance thinking ability. DeepSeek-R1 attains results on par with [OpenAI's](https://smaphofilm.com) o1 model on a number of criteria, including MATH-500 and SWE-bench.<br> |
||||
|
<br>DeepSeek-R1 is based on DeepSeek-V3, a mixture of specialists (MoE) design recently open-sourced by DeepSeek. This base model is [fine-tuned](https://social.updum.com) [utilizing](http://gitlab.together.social) Group Relative Policy Optimization (GRPO), a [reasoning-oriented variation](https://axeplex.com) of RL. The research study group also [performed knowledge](https://edu.shpl.ru) distillation from DeepSeek-R1 to [open-source Qwen](http://www.todak.co.kr) and Llama models and launched numerous variations of each |
Loading…
Reference in new issue