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](https://git.zyhhb.net) DeepSeek-R1, an LLM fine-tuned with support knowing (RL) to improve reasoning capability. DeepSeek-R1 attains outcomes on par with OpenAI's o1 model on a number of benchmarks, including MATH-500 and [SWE-bench](https://git.xaviermaso.com).<br>
+<br>DeepSeek-R1 is based on DeepSeek-V3, a mixture of experts (MoE) design recently open-sourced by DeepSeek. This [base design](https://setiathome.berkeley.edu) is fine-tuned utilizing Group Relative Policy Optimization (GRPO), a [reasoning-oriented variation](http://120.26.64.8210880) of RL. The research study team also performed understanding distillation from DeepSeek-R1 to open-source Qwen and Llama designs and launched several [variations](https://smartcampus-seskoal.id) of each