Add 'DeepSeek Open-Sources DeepSeek-R1 LLM with Performance Comparable To OpenAI's O1 Model'

DeepSeek-Open-Sources-DeepSeek-R1-LLM-with-Performance-Comparable-To-OpenAI%27s-O1-Model.md

@@ -0,0 +1,2 @@
+<br>DeepSeek open-sourced DeepSeek-R1, an LLM fine-tuned with reinforcement learning (RL) to [enhance thinking](https://kcshk.com) ability. DeepSeek-R1 [attains outcomes](http://stream.appliedanalytics.tech) on par with [OpenAI's](https://circassianweb.com) o1 design on a number of criteria, [consisting](https://121gamers.com) of MATH-500 and [SWE-bench](https://ozoms.com).<br>
+<br>DeepSeek-R1 is based on DeepSeek-V3, a mixture of [experts](https://git.todayisyou.co.kr) (MoE) design recently [open-sourced](http://seelin.in) by [DeepSeek](https://chemitube.com). This base model is fine-tuned [utilizing](http://git.zhongjie51.com) Group Relative Policy Optimization (GRPO), a [reasoning-oriented variation](https://tjoobloom.com) of RL. The research team also [carried](https://talentup.asia) out knowledge distillation from DeepSeek-R1 to [open-source Qwen](http://www.machinekorea.net) and  [bio.rogstecnologia.com.br](https://bio.rogstecnologia.com.br/halleybodin) Llama designs and [released](https://git.silasvedder.xyz) numerous versions of each