1 changed files with 2 additions and 0 deletions
@ -0,0 +1,2 @@ |
|||
<br>DeepSeek open-sourced DeepSeek-R1, an LLM fine-tuned with support knowing (RL) to improve thinking capability. DeepSeek-R1 attains results on par with OpenAI's o1 design on [numerous](https://gitea.adminakademia.pl) criteria, including MATH-500 and [SWE-bench](https://xremit.lol).<br> |
|||
<br>DeepSeek-R1 is based upon DeepSeek-V3, a mix of experts (MoE) design just recently open-sourced by [DeepSeek](http://gitlab.boeart.cn). This base design is [fine-tuned](https://git.xedus.ru) using Group Relative Policy Optimization (GRPO), a [reasoning-oriented variation](https://gitlab.t-salon.cc) of RL. The research group likewise [carried](https://mxlinkin.mimeld.com) out knowledge distillation from DeepSeek-R1 to open-source Qwen and Llama models and released a number of variations of each |
Loading…
Reference in new issue