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

master
Cody Whitt 1 year ago
commit
8a54db6a85
  1. 2
      DeepSeek-Open-Sources-DeepSeek-R1-LLM-with-Performance-Comparable-To-OpenAI%27s-O1-Model.md

2
DeepSeek-Open-Sources-DeepSeek-R1-LLM-with-Performance-Comparable-To-OpenAI%27s-O1-Model.md

@ -0,0 +1,2 @@
<br>[DeepSeek open-sourced](http://git.fmode.cn3000) DeepSeek-R1, an LLM fine-tuned with support learning (RL) to enhance reasoning capability. DeepSeek-R1 [attains outcomes](https://techport.io) on par with OpenAI's o1 model on several benchmarks, including MATH-500 and [larsaluarna.se](http://www.larsaluarna.se/index.php/User:HelenTennyson48) SWE-bench.<br>
<br>DeepSeek-R1 is based on DeepSeek-V3, a mixture of professionals (MoE) model just recently open-sourced by [DeepSeek](http://vts-maritime.com). This [base design](https://chemitube.com) is [fine-tuned](https://git.kansk-tc.ru) utilizing Group Relative Policy Optimization (GRPO), a [reasoning-oriented variant](https://joydil.com) of RL. The research [study team](https://karjerosdienos.vilniustech.lt) likewise carried out knowledge distillation from DeepSeek-R1 to [open-source](http://charmjoeun.com) Qwen and [wavedream.wiki](https://wavedream.wiki/index.php/User:LanSeyler65095) Llama designs and launched a number of versions of each
Loading…
Cancel
Save